Conference on Empirical Methods in Natural Language Processing (2022)



bib (full) Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

pdf bib
Generative Knowledge Graph Construction: A Review
Hongbin Ye | Ningyu Zhang | Hui Chen | Huajun Chen

Generative Knowledge Graph Construction (KGC) refers to those methods that leverage the sequence-to-sequence framework for building knowledge graphs, which is flexible and can be adapted to widespread tasks. In this study, we summarize the recent compelling progress in generative knowledge graph construction. We present the advantages and weaknesses of each paradigm in terms of different generation targets and provide theoretical insight and empirical analysis. Based on the review, we suggest promising research directions for the future. Our contributions are threefold: (1) We present a detailed, complete taxonomy for the generative KGC methods; (2) We provide a theoretical and empirical analysis of the generative KGC methods; (3) We propose several research directions that can be developed in the future.

pdf bib
CDConv: A Benchmark for Contradiction Detection in Chinese Conversations
Chujie Zheng | Jinfeng Zhou | Yinhe Zheng | Libiao Peng | Zhen Guo | Wenquan Wu | Zheng-Yu Niu | Hua Wu | Minlie Huang

Dialogue contradiction is a critical issue in open-domain dialogue systems. The contextualization nature of conversations makes dialogue contradiction detection rather challenging. In this work, we propose a benchmark for Contradiction Detection in Chinese Conversations, namely CDConv. It contains 12K multi-turn conversations annotated with three typical contradiction categories: Intra-sentence Contradiction, Role Confusion, and History Contradiction. To efficiently construct the CDConv conversations, we devise a series of methods for automatic conversation generation, which simulate common user behaviors that trigger chatbots to make contradictions. We conduct careful manual quality screening of the constructed conversations and show that state-of-the-art Chinese chatbots can be easily goaded into making contradictions. Experiments on CDConv show that properly modeling contextual information is critical for dialogue contradiction detection, but there are still unresolved challenges that require future research.

pdf bib
Transformer Feed-Forward Layers Build Predictions by Promoting Concepts in the Vocabulary Space
Mor Geva | Avi Caciularu | Kevin Wang | Yoav Goldberg

Transformer-based language models (LMs) are at the core of modern NLP, but their internal prediction construction process is opaque and largely not understood. In this work, we make a substantial step towards unveiling this underlying prediction process, by reverse-engineering the operation of the feed-forward network (FFN) layers, one of the building blocks of transformer models. We view the token representation as a changing distribution over the vocabulary, and the output from each FFN layer as an additive update to that distribution. Then, we analyze the FFN updates in the vocabulary space, showing that each update can be decomposed to sub-updates corresponding to single FFN parameter vectors, each promoting concepts that are often human-interpretable. We then leverage these findings for controlling LM predictions, where we reduce the toxicity of GPT2 by almost 50%, and for improving computation efficiency with a simple early exit rule, saving 20% of computation on average.

Learning to Generate Question by Asking Question: A Primal-Dual Approach with Uncommon Word Generation
Qifan Wang | Li Yang | Xiaojun Quan | Fuli Feng | Dongfang Liu | Zenglin Xu | Sinong Wang | Hao Ma

Automatic question generation (AQG) is the task of generating a question from a given passage and an answer. Most existing AQG methods aim at encoding the passage and the answer to generate the question. However, limited work has focused on modeling the correlation between the target answer and the generated question. Moreover, unseen or rare word generation has not been studied in previous works. In this paper, we propose a novel approach which incorporates question generation with its dual problem, question answering, into a unified primal-dual framework. Specifically, the question generation component consists of an encoder that jointly encodes the answer with the passage, and a decoder that produces the question. The question answering component then re-asks the generated question on the passage to ensure that the target answer is obtained. We further introduce a knowledge distillation module to improve the model generalization ability. We conduct an extensive set of experiments on SQuAD and HotpotQA benchmarks. Experimental results demonstrate the superior performance of the proposed approach over several state-of-the-art methods.

Graph-based Model Generation for Few-Shot Relation Extraction
Wanli Li | Tieyun Qian

Few-shot relation extraction (FSRE) has been a challenging problem since it only has a handful of training instances. Existing models follow a ‘one-for-all’ scheme where one general large model performs all individual N-way-K-shot tasks in FSRE, which prevents the model from achieving the optimal point on each task. In view of this, we propose a model generation framework that consists of one general model for all tasks and many tiny task-specific models for each individual task. The general model generates and passes the universal knowledge to the tiny models which will be further fine-tuned when performing specific tasks. In this way, we decouple the complexity of the entire task space from that of all individual tasks while absorbing the universal knowledge.Extensive experimental results on two public datasets demonstrate that our framework reaches a new state-of-the-art performance for FRSE tasks. Our code is available at:

Backdoor Attacks in Federated Learning by Rare Embeddings and Gradient Ensembling
Ki Yoon Yoo | Nojun Kwak

Recent advances in federated learning have demonstrated its promising capability to learn on decentralized datasets. However, a considerable amount of work has raised concerns due to the potential risks of adversaries participating in the framework to poison the global model for an adversarial purpose. This paper investigates the feasibility of model poisoning for backdoor attacks through rare word embeddings of NLP models. In text classification, less than 1% of adversary clients suffices to manipulate the model output without any drop in the performance of clean sentences. For a less complex dataset, a mere 0.1% of adversary clients is enough to poison the global model effectively. We also propose a technique specialized in the federated learning scheme called gradient ensemble, which enhances the backdoor performance in all experimental settings.

Generating Natural Language Proofs with Verifier-Guided Search
Kaiyu Yang | Jia Deng | Danqi Chen

Reasoning over natural language is a challenging problem in NLP. In this work, we focus on proof generation: Given a hypothesis and a set of supporting facts, the model generates a proof tree indicating how to derive the hypothesis from supporting facts. Compared to generating the entire proof in one shot, stepwise generation can better exploit the compositionality and generalize to longer proofs but has achieved limited success on real-world data. Existing stepwise methods struggle to generate proof steps that are both logically valid and relevant to the hypothesis. Instead, they tend to hallucinate invalid steps given the hypothesis. In this paper, we present a novel stepwise method, NLProofS (Natural Language Proof Search), which learns to generate relevant steps conditioning on the hypothesis. At the core of our approach, we train an independent verifier to check the validity of the proof steps to prevent hallucination. Instead of generating steps greedily, we search for proofs maximizing a global proof score judged by the verifier. NLProofS achieves state-of-the-art performance on EntailmentBank and RuleTaker. Specifically, it improves the correctness of predicted proofs from 27.7% to 33.3% in the distractor setting of EntailmentBank, demonstrating the effectiveness of NLProofS in generating challenging human-authored proofs.

Toward Unifying Text Segmentation and Long Document Summarization
Sangwoo Cho | Kaiqiang Song | Xiaoyang Wang | Fei Liu | Dong Yu

Text segmentation is important for signaling a document’s structure. Without segmenting a long document into topically coherent sections, it is difficult for readers to comprehend the text, let alone find important information. The problem is only exacerbated by a lack of segmentation in transcripts of audio/video recordings. In this paper, we explore the role that section segmentation plays in extractive summarization of written and spoken documents. Our approach learns robust sentence representations by performing summarization and segmentation simultaneously, which is further enhanced by an optimization-based regularizer to promote selection of diverse summary sentences. We conduct experiments on multiple datasets ranging from scientific articles to spoken transcripts to evaluate the model’s performance. Our findings suggest that the model can not only achieve state-of-the-art performance on publicly available benchmarks, but demonstrate better cross-genre transferability when equipped with text segmentation. We perform a series of analyses to quantify the impact of section segmentation on summarizing written and spoken documents of substantial length and complexity.

The Geometry of Multilingual Language Model Representations
Tyler Chang | Zhuowen Tu | Benjamin Bergen

We assess how multilingual language models maintain a shared multilingual representation space while still encoding language-sensitive information in each language. Using XLM-R as a case study, we show that languages occupy similar linear subspaces after mean-centering, evaluated based on causal effects on language modeling performance and direct comparisons between subspaces for 88 languages. The subspace means differ along language-sensitive axes that are relatively stable throughout middle layers, and these axes encode information such as token vocabularies. Shifting representations by language means is sufficient to induce token predictions in different languages. However, we also identify stable language-neutral axes that encode information such as token positions and part-of-speech. We visualize representations projected onto language-sensitive and language-neutral axes, identifying language family and part-of-speech clusters, along with spirals, toruses, and curves representing token position information. These results demonstrate that multilingual language models encode information along orthogonal language-sensitive and language-neutral axes, allowing the models to extract a variety of features for downstream tasks and cross-lingual transfer learning.

Improving Complex Knowledge Base Question Answering via Question-to-Action and Question-to-Question Alignment
Yechun Tang | Xiaoxia Cheng | Weiming Lu

Complex knowledge base question answering can be achieved by converting questions into sequences of predefined actions. However, there is a significant semantic and structural gap between natural language and action sequences, which makes this conversion difficult. In this paper, we introduce an alignment-enhanced complex question answering framework, called ALCQA, which mitigates this gap through question-to-action alignment and question-to-question alignment. We train a question rewriting model to align the question and each action, and utilize a pretrained language model to implicitly align the question and KG artifacts. Moreover, considering that similar questions correspond to similar action sequences, we retrieve top-k similar question-answer pairs at the inference stage through question-to-question alignment and propose a novel reward-guided action sequence selection strategy to select from candidate action sequences. We conduct experiments on CQA and WQSP datasets, and the results show that our approach outperforms state-of-the-art methods and obtains a 9.88% improvements in the F1 metric on CQA dataset. Our source code is available at

PAIR: Prompt-Aware margIn Ranking for Counselor Reflection Scoring in Motivational Interviewing
Do June Min | Verónica Pérez-Rosas | Kenneth Resnicow | Rada Mihalcea

Counselor reflection is a core verbal skill used by mental health counselors to express understanding and affirmation of the client’s experience and concerns. In this paper, we propose a system for the analysis of counselor reflections. Specifically, our system takes as input one dialog turn containing a client prompt and a counselor response, and outputs a score indicating the level of reflection in the counselor response. We compile a dataset consisting of different levels of reflective listening skills, and propose the Prompt-Aware margIn Ranking (PAIR) framework that contrasts positive and negative prompt and response pairs using specially designed multi-gap and prompt-aware margin ranking losses. Through empirical evaluations and deployment of our system in a real-life educational environment, we show that our analysis model outperforms several baselines on different metrics, and can be used to provide useful feedback to counseling trainees.

Co-guiding Net: Achieving Mutual Guidances between Multiple Intent Detection and Slot Filling via Heterogeneous Semantics-Label Graphs
Bowen Xing | Ivor Tsang

Recent graph-based models for joint multiple intent detection and slot filling have obtained promising results through modeling the guidance from the prediction of intents to the decoding of slot filling.However, existing methods (1) only model the unidirectional guidance from intent to slot; (2) adopt homogeneous graphs to model the interactions between the slot semantics nodes and intent label nodes, which limit the performance.In this paper, we propose a novel model termed Co-guiding Net, which implements a two-stage framework achieving the mutual guidances between the two tasks.In the first stage, the initial estimated labels of both tasks are produced, and then they are leveraged in the second stage to model the mutual guidances.Specifically, we propose two heterogeneous graph attention networks working on the proposed two heterogeneous semantics-label graphs, which effectively represent the relations among the semantics nodes and label nodes.Experiment results show that our model outperforms existing models by a large margin, obtaining a relative improvement of 19.3% over the previous best model on MixATIS dataset in overall accuracy.

The Importance of Being Parameters: An Intra-Distillation Method for Serious Gains
Haoran Xu | Philipp Koehn | Kenton Murray

Recent model pruning methods have demonstrated the ability to remove redundant parameters without sacrificing model performance. Common methods remove redundant parameters according to the parameter sensitivity, a gradient-based measure reflecting the contribution of the parameters. In this paper, however, we argue that redundant parameters can be trained to make beneficial contributions. We first highlight the large sensitivity (contribution) gap among high-sensitivity and low-sensitivity parameters and show that the model generalization performance can be significantly improved after balancing the contribution of all parameters. Our goal is to balance the sensitivity of all parameters and encourage all of them to contribute equally. We propose a general task-agnostic method, namely intra-distillation, appended to the regular training loss to balance parameter sensitivity. Moreover, we also design a novel adaptive learning method to control the strength of intra-distillation loss for faster convergence. Our experiments show the strong effectiveness of our methods on machine translation, natural language understanding, and zero-shot cross-lingual transfer across up to 48 languages, e.g., a gain of 3.54 BLEU on average across 8 language pairs from the IWSLT’14 dataset.

Interpreting Language Models with Contrastive Explanations
Kayo Yin | Graham Neubig

Model interpretability methods are often used to explain NLP model decisions on tasks such as text classification, where the output space is relatively small. However, when applied to language generation, where the output space often consists of tens of thousands of tokens, these methods are unable to provide informative explanations. Language models must consider various features to predict a token, such as its part of speech, number, tense, or semantics.Existing explanation methods conflate evidence for all these features into a single explanation, which is less interpretable for human understanding.To disentangle the different decisions in language modeling, we focus on explaining language models contrastively: we look for salient input tokens that explain why the model predicted one token instead of another. We demonstrate that contrastive explanations are quantifiably better than non-contrastive explanations in verifying major grammatical phenomena, and that they significantly improve contrastive model simulatability for human observers. We also identify groups of contrastive decisions where the model uses similar evidence, and we are able to characterize what input tokens models use during various language generation decisions.

RankGen: Improving Text Generation with Large Ranking Models
Kalpesh Krishna | Yapei Chang | John Wieting | Mohit Iyyer

Given an input sequence (or prefix), modern language models often assign high probabilities to output sequences that are repetitive, incoherent, or irrelevant to the prefix; as such, model-generated text also contains such artifacts. To address these issues we present RankGen, a 1.2B parameter encoder model for English that scores model generations given a prefix. RankGen can be flexibly incorporated as a scoring function in beam search and used to decode from any pretrained language model. We train RankGen using large-scale contrastive learning to map a prefix close to the ground-truth sequence that follows it and far away from two types of negatives: (1) random sequences from the same document as the prefix, and (2) sequences generated from a large language model conditioned on the prefix. Experiments across four different language models (345M-11B parameters) and two domains show that RankGen significantly outperforms decoding algorithms like nucleus, top-k, and typical sampling on both automatic metrics (85.0 vs 77.3 MAUVE) as well as human evaluations with English writers (74.5% human preference over nucleus sampling). Analysis reveals that RankGen outputs are more relevant to the prefix and improve continuity and coherence compared to baselines. We release our model checkpoints, code, and human preference data with explanations to facilitate future research.

Learning a Grammar Inducer from Massive Uncurated Instructional Videos
Songyang Zhang | Linfeng Song | Lifeng Jin | Haitao Mi | Kun Xu | Dong Yu | Jiebo Luo

Video-aided grammar induction aims to leverage video information for finding more accurate syntactic grammars for accompanying text. While previous work focuses on building systems for inducing grammars on text that are well-aligned with video content, we investigate the scenario, in which text and video are only in loose correspondence. Such data can be found in abundance online, and the weak correspondence is similar to the indeterminacy problem studied in language acquisition. Furthermore, we build a new model that can better learn video-span correlation without manually designed features adopted by previous work. Experiments show that our model trained only on large-scale YouTube data with no text-video alignment reports strong and robust performances across three unseen datasets, despite domain shift and noisy label issues. Furthermore our model yields higher F1 scores than the previous state-of-the-art systems trained on in-domain data.

Normalized Contrastive Learning for Text-Video Retrieval
Yookoon Park | Mahmoud Azab | Seungwhan Moon | Bo Xiong | Florian Metze | Gourab Kundu | Kirmani Ahmed

Cross-modal contrastive learning has led the recent advances in multimodal retrieval with its simplicity and effectiveness. In this work, however, we reveal that cross-modal contrastive learning suffers from incorrect normalization of the sum retrieval probabilities of each text or video instance. Specifically, we show that many test instances are either over- or under-represented during retrieval, significantly hurting the retrieval performance. To address this problem, we propose Normalized Contrastive Learning (NCL) which utilizes the Sinkhorn-Knopp algorithm to compute the instance-wise biases that properly normalize the sum retrieval probabilities of each instance so that every text and video instance is fairly represented during cross-modal retrieval. Empirical study shows that NCL brings consistent and significant gains in text-video retrieval on different model architectures, with new state-of-the-art multimodal retrieval metrics on the ActivityNet, MSVD, and MSR-VTT datasets without any architecture engineering.

Estimating Soft Labels for Out-of-Domain Intent Detection
Hao Lang | Yinhe Zheng | Jian Sun | Fei Huang | Luo Si | Yongbin Li

Out-of-Domain (OOD) intent detection is important for practical dialog systems. To alleviate the issue of lacking OOD training samples, some works propose synthesizing pseudo OOD samples and directly assigning one-hot OOD labels to these pseudo samples. However, these one-hot labels introduce noises to the training process because some “hard” pseudo OOD samples may coincide with In-Domain (IND) intents. In this paper, we propose an adaptive soft pseudo labeling (ASoul) method that can estimate soft labels for pseudo OOD samples when training OOD detectors. Semantic connections between pseudo OOD samples and IND intents are captured using an embedding graph. A co-training framework is further introduced to produce resulting soft labels following the smoothness assumption, i.e., close samples are likely to have similar labels. Extensive experiments on three benchmark datasets show that ASoul consistently improves the OOD detection performance and outperforms various competitive baselines.

Multi-VQG: Generating Engaging Questions for Multiple Images
Min-Hsuan Yeh | Vincent Chen | Ting-Hao Huang | Lun-Wei Ku

Generating engaging content has drawn much recent attention in the NLP community. Asking questions is a natural way to respond to photos and promote awareness. However, most answers to questions in traditional question-answering (QA) datasets are factoids, which reduce individuals’ willingness to answer. Furthermore, traditional visual question generation (VQG) confines the source data for question generation to single images, resulting in a limited ability to comprehend time-series information of the underlying event. In this paper, we propose generating engaging questions from multiple images. We present MVQG, a new dataset, and establish a series of baselines, including both end-to-end and dual-stage architectures. Results show that building stories behind the image sequence enables models togenerate engaging questions, which confirms our assumption that people typically construct a picture of the event in their minds before asking questions. These results open up an exciting challenge for visual-and-language models to implicitly construct a story behind a series of photos to allow for creativity and experience sharing and hence draw attention to downstream applications.

Tomayto, Tomahto. Beyond Token-level Answer Equivalence for Question Answering Evaluation
Jannis Bulian | Christian Buck | Wojciech Gajewski | Benjamin Börschinger | Tal Schuster

The predictions of question answering (QA) systems are typically evaluated against manually annotated finite sets of one or more answers. This leads to a coverage limitation that results in underestimating the true performance of systems, and is typically addressed by extending over exact match (EM) with predefined rules or with the token-level F1 measure.In this paper, we present the first systematic conceptual and data-driven analysis to examine the shortcomings of token-level equivalence measures.To this end, we define the asymmetric notion of answer equivalence (AE), accepting answers that are equivalent to or improve over the reference, and publish over 23k human judgements for candidates produced by multiple QA systems on SQuAD.Through a careful analysis of this data, we reveal and quantify several concrete limitations of the F1 measure, such as a false impression of graduality, or missing dependence on the question.Since collecting AE annotations for each evaluated model is expensive, we learn a BERT matching (BEM) measure to approximate this task. Being a simpler task than QA, we find BEM to provide significantly better AE approximations than F1, and to more accurately reflect the performance of systems.Finally, we demonstrate the practical utility of AE and BEM on the concrete application of minimal accurate prediction sets, reducing the number of required answers by up to X2.6.

Non-Parametric Domain Adaptation for End-to-End Speech Translation
Yichao Du | Weizhi Wang | Zhirui Zhang | Boxing Chen | Tong Xu | Jun Xie | Enhong Chen

The end-to-end speech translation (E2E-ST) has received increasing attention due to the potential of its less error propagation, lower latency and fewer parameters. However, the effectiveness of neural-based approaches to this task is severely limited by the available training corpus, especially for domain adaptation where in-domain triplet data is scarce or nonexistent. In this paper, we propose a novel non-parametric method that leverages in-domain text translation corpus to achieve domain adaptation for E2E-ST systems. To this end, we first incorporate an additional encoder into the pre-trained E2E-ST model to realize text translation modeling, based on which the decoder’s output representations for text and speech translation tasks are unified by reducing the correspondent representation mismatch in available triplet training data. During domain adaptation, a k-nearest-neighbor (kNN) classifier is introduced to produce the final translation distribution using the external datastore built by the domain-specific text translation corpus, while the universal output representation is adopted to perform a similarity search. Experiments on the Europarl-ST benchmark demonstrate that when in-domain text translation data is involved only, our proposed approach significantly improves baseline by 12.82 BLEU on average in all translation directions, even outperforming the strong in-domain fine-tuning strategy.

Prompting for Multimodal Hateful Meme Classification
Rui Cao | Roy Ka-Wei Lee | Wen-Haw Chong | Jing Jiang

Hateful meme classification is a challenging multimodal task that requires complex reasoning and contextual background knowledge. Ideally, we could leverage an explicit external knowledge base to supplement contextual and cultural information in hateful memes. However, there is no known explicit external knowledge base that could provide such hate speech contextual information. To address this gap, we propose PromptHate, a simple yet effective prompt-based model that prompts pre-trained language models (PLMs) for hateful meme classification. Specifically, we construct simple prompts and provide a few in-context examples to exploit the implicit knowledge in the pre-trained RoBERTa language model for hateful meme classification. We conduct extensive experiments on two publicly available hateful and offensive meme datasets. Our experiment results show that PromptHate is able to achieve a high AUC of 90.96, outperforming state-of-the-art baselines on the hateful meme classification task. We also perform fine-grain analyses and case studies on various prompt settings and demonstrate the effectiveness of the prompts on hateful meme classification.

Certified Error Control of Candidate Set Pruning for Two-Stage Relevance Ranking
Minghan Li | Xinyu Zhang | Ji Xin | Hongyang Zhang | Jimmy Lin

In information retrieval (IR), candidate set pruning has been commonly used to speed up two-stage relevance ranking. However, such an approach lacks accurate error control and often trades accuracy against computational efficiency in an empirical fashion, missing theoretical guarantees. In this paper, we propose the concept of certified error control of candidate set pruning for relevance ranking, which means that the test error after pruning is guaranteed to be controlled under a user-specified threshold with high probability. Both in-domain and out-of-domain experiments show that our method successfully prunes the first-stage retrieved candidate sets to improve the second-stage reranking speed while satisfying the pre-specified accuracy constraints in both settings. For example, on MS MARCO Passage v1, our method reduces the average candidate set size from 1000 to 27, increasing reranking speed by about 37 times, while keeping MRR@10 greater than a pre-specified value of 0.38 with about 90% empirical coverage. In contrast, empirical baselines fail to meet such requirements. Code and data are available at:

Linearizing Transformer with Key-Value Memory
Yizhe Zhang | Deng Cai

Efficient transformer variants with linear time complexity have been developed to mitigate the quadratic computational overhead of the vanilla transformer. Among them are low-rank projection methods such as Linformer and kernel-based Transformers. Despite their unique merits, they usually suffer from a performance drop comparing with the vanilla transformer on many sequence generation tasks, and often fail to obtain computation gain when the generation is short. We propose Memsizer, an approach towards closing the performance gap while improving the efficiency even with short generation. It projects the source sequences into lower dimension representations like Linformer, while enjoying efficient recurrent-style incremental computation similar to kernel-based transformers. This yields linear computation time and constant memory complexity at inference time. Memsizer also employs a lightweight multi-head mechanism which renders the computation as light as a single-head model. We demonstrate that Memsizer provides an improved balance between efficiency and accuracy over the vanilla transformer and other efficient transformer variants in three typical sequence generation tasks, including machine translation, abstractive text summarization, and language modeling.

Robustness of Fusion-based Multimodal Classifiers to Cross-Modal Content Dilutions
Gaurav Verma | Vishwa Vinay | Ryan Rossi | Srijan Kumar

As multimodal learning finds applications in a wide variety of high-stakes societal tasks, investigating their robustness becomes important. Existing work has focused on understanding the robustness of vision-and-language models to imperceptible variations on benchmark tasks. In this work, we investigate the robustness of multimodal classifiers to cross-modal dilutions – a plausible variation. We develop a model that, given a multimodal (image + text) input, generates additional dilution text that (a) maintains relevance and topical coherence with the image and existing text, and (b) when added to the original text, leads to misclassification of the multimodal input. Via experiments on Crisis Humanitarianism and Sentiment Detection tasks, we find that the performance of task-specific fusion-based multimodal classifiers drops by 23.3% and 22.5%, respectively, in the presence of dilutions generated by our model. Metric-based comparisons with several baselines and human evaluations indicate that our dilutions show higher relevance and topical coherence, while simultaneously being more effective at demonstrating the brittleness of the multimodal classifiers. Our work aims to highlight and encourage further research on the robustness of deep multimodal models to realistic variations, especially in human-facing societal applications.

Translation between Molecules and Natural Language
Carl Edwards | Tuan Lai | Kevin Ros | Garrett Honke | Kyunghyun Cho | Heng Ji

We present MolT5 - a self-supervised learning framework for pretraining models on a vast amount of unlabeled natural language text and molecule strings. MolT5 allows for new, useful, and challenging analogs of traditional vision-language tasks, such as molecule captioning and text-based de novo molecule generation (altogether: translation between molecules and language), which we explore for the first time. Since MolT5 pretrains models on single-modal data, it helps overcome the chemistry domain shortcoming of data scarcity. Furthermore, we consider several metrics, including a new cross-modal embedding-based metric, to evaluate the tasks of molecule captioning and text-based molecule generation. Our results show that MolT5-based models are able to generate outputs, both molecules and captions, which in many cases are high quality.

What Makes Instruction Learning Hard? An Investigation and a New Challenge in a Synthetic Environment
Matthew Finlayson | Kyle Richardson | Ashish Sabharwal | Peter Clark

The instruction learning paradigm—where a model learns to perform new tasks from task descriptions alone—has become popular in research on general-purpose models. The capabilities of large transformer models as instruction learners, however, remain poorly understood. We use a controlled synthetic environment to characterize such capabilities. Specifically, we use the task of deciding whether a given string matches a regular expression (viewed as an instruction) to identify properties of tasks, instructions, and instances that make instruction learning challenging. For instance, we find that our model, a fine-tuned T5-based text2text transformer, struggles with large regular languages, suggesting that less precise instructions are challenging for models. Instruction executions that require tracking longer contexts of prior steps are also difficult. We use our findings to systematically construct a challenging instruction learning dataset, which we call Hard RegSet. Fine-tuning on Hard RegSet, our large transformer learns to correctly interpret (with at least 90% accuracy) only 65.6% of test instructions, and 11%-24% of the instructions in out-of-distribution generalization settings. We thus propose Hard RegSet as a challenging instruction learning dataset, and a controlled environment for studying instruction learning.

Sentence-Incremental Neural Coreference Resolution
Matt Grenander | Shay B. Cohen | Mark Steedman

We propose a sentence-incremental neural coreference resolution system which incrementally builds clusters after marking mention boundaries in a shift-reduce method. The system is aimed at bridging two recent approaches at coreference resolution: (1) state-of-the-art non-incremental models that incur quadratic complexity in document length with high computational cost, and (2) memory network-based models which operate incrementally but do not generalize beyond pronouns. For comparison, we simulate an incremental setting by constraining non-incremental systems to form partial coreference chains before observing new sentences. In this setting, our system outperforms comparable state-of-the-art methods by 2 F1 on OntoNotes and 6.8 F1 on the CODI-CRAC 2021 corpus. In a conventional coreference setup, our system achieves 76.3 F1 on OntoNotes and 45.5 F1 on CODI-CRAC 2021, which is comparable to state-of-the-art baselines. We also analyze variations of our system and show that the degree of incrementality in the encoder has a surprisingly large effect on the resulting performance.

SNaC: Coherence Error Detection for Narrative Summarization
Tanya Goyal | Junyi Jessy Li | Greg Durrett

Progress in summarizing long texts is inhibited by the lack of appropriate evaluation frameworks. A long summary that appropriately covers the facets of that text must also present a coherent narrative, but current automatic and human evaluation methods fail to identify gaps in coherence. In this work, we introduce SNaC, a narrative coherence evaluation framework for fine-grained annotations of long summaries. We develop a taxonomy of coherence errors in generated narrative summaries and collect span-level annotations for 6.6k sentences across 150 book and movie summaries. Our work provides the first characterization of coherence errors generated by state-of-the-art summarization models and a protocol for eliciting coherence judgments from crowdworkers. Furthermore, we show that the collected annotations allow us to benchmark past work in coherence modeling and train a strong classifier for automatically localizing coherence errors in generated summaries. Finally, our SNaC framework can support future work in long document summarization and coherence evaluation, including improved summarization modeling and post-hoc summary correction.

HydraSum: Disentangling Style Features in Text Summarization with Multi-Decoder Models
Tanya Goyal | Nazneen Rajani | Wenhao Liu | Wojciech Kryscinski

Summarization systems make numerous “decisions” about summary properties during inference, e.g. degree of copying, specificity and length of outputs, etc. However, these are implicitly encoded within model parameters and specific styles cannot be enforced. To address this, we introduce HydraSum, a new summarization architecture that extends the single decoder framework of current models to a mixture-of-experts version with multiple decoders. We show that HydraSum’s multiple decoders automatically learn contrasting summary styles when trained under the standard training objective without any extra supervision. Through experiments on three summarization datasets (CNN, Newsroom and XSum), we show that HydraSum provides a simple mechanism to obtain stylistically-diverse summaries by sampling from either individual decoders or their mixtures, outperforming baseline models. Finally, we demonstrate that a small modification to the gating strategy during training can enforce an even stricter style partitioning, e.g. high- vs low-abstractiveness or high- vs low-specificity, allowing users to sample from a larger area in the generation space and vary summary styles along multiple dimensions.

A Good Neighbor, A Found Treasure: Mining Treasured Neighbors for Knowledge Graph Entity Typing
Zhuoran Jin | Pengfei Cao | Yubo Chen | Kang Liu | Jun Zhao

The task of knowledge graph entity typing (KGET) aims to infer the missing types for entities in knowledge graphs. Some pioneering work has proved that neighbor information is very important for the task. However, existing methods only leverage the one-hop neighbor information of the central entity, ignoring the multi-hop neighbor information that can provide valuable clues for inference. Besides, we also observe that there are co-occurrence relations between types, which is very helpful to alleviate false-negative problem. In this paper, we propose a novel method called Mining Treasured Neighbors (MiNer) to make use of these two characteristics. Firstly, we devise a Neighbor Information Aggregation module to aggregate the neighbor information. Then, we propose an Entity Type Inference module to mitigate the adverse impact of the irrelevant neighbor information. Finally, a Type Co-occurrence Regularization module is designed to prevent the model from overfitting the false negative examples caused by missing types. Experimental results on two widely used datasets indicate that our approach significantly outperforms previous state-of-the-art methods.

Guiding Neural Entity Alignment with Compatibility
Bing Liu | Harrisen Scells | Wen Hua | Guido Zuccon | Genghong Zhao | Xia Zhang

Entity Alignment (EA) aims to find equivalent entities between two Knowledge Graphs (KGs). While numerous neural EA models have been devised, they are mainly learned using labelled data only. In this work, we argue that different entities within one KG should have compatible counterparts in the other KG due to the potential dependencies among the entities. Making compatible predictions thus should be one of the goals of training an EA model along with fitting the labelled data: this aspect however is neglected in current methods. To power neural EA models with compatibility, we devise a training framework by addressing three problems: (1) how to measure the compatibility of an EA model; (2) how to inject the property of being compatible into an EA model; (3) how to optimise parameters of the compatibility model. Extensive experiments on widely-used datasets demonstrate the advantages of integrating compatibility within EA models. In fact, state-of-the-art neural EA models trained within our framework using just 5% of the labelled data can achieve comparable effectiveness with supervised training using 20% of the labelled data.

InstructDial: Improving Zero and Few-shot Generalization in Dialogue through Instruction Tuning
Prakhar Gupta | Cathy Jiao | Yi-Ting Yeh | Shikib Mehri | Maxine Eskenazi | Jeffrey Bigham

Instruction tuning is an emergent paradigm in NLP wherein natural language instructions are leveraged with language models to induce zero-shot performance on unseen tasks. Dialogue is an especially interesting area in which to explore instruction tuning because dialogue systems perform multiple kinds of tasks related to language (e.g., natural language understanding and generation, domain-specific interaction), yet instruction tuning has not been systematically explored for dialogue-related tasks. We introduce InstructDial, an instruction tuning framework for dialogue, which consists of a repository of 48 diverse dialogue tasks in a unified text-to-text format created from 59 openly available dialogue datasets. We explore cross-task generalization ability on models tuned on InstructDial across diverse dialogue tasks. Our analysis reveals that InstructDial enables good zero-shot performance on unseen datasets and tasks such as dialogue evaluation and intent detection, and even better performance in a few-shot setting. To ensure that models adhere to instructions, we introduce novel meta-tasks. We establish benchmark zero-shot and few-shot performance of models trained using the proposed framework on multiple dialogue tasks.

Unsupervised Boundary-Aware Language Model Pretraining for Chinese Sequence Labeling
Peijie Jiang | Dingkun Long | Yanzhao Zhang | Pengjun Xie | Meishan Zhang | Min Zhang

Boundary information is critical for various Chinese language processing tasks, such as word segmentation, part-of-speech tagging, and named entity recognition. Previous studies usually resorted to the use of a high-quality external lexicon, where lexicon items can offer explicit boundary information. However, to ensure the quality of the lexicon, great human effort is always necessary, which has been generally ignored. In this work, we suggest unsupervised statistical boundary information instead, and propose an architecture to encode the information directly into pre-trained language models, resulting in Boundary-Aware BERT (BABERT). We apply BABERT for feature induction of Chinese sequence labeling tasks. Experimental results on ten benchmarks of Chinese sequence labeling demonstrate that BABERT can provide consistent improvements on all datasets. In addition, our method can complement previous supervised lexicon exploration, where further improvements can be achieved when integrated with external lexicon information.

RetroMAE: Pre-Training Retrieval-oriented Language Models Via Masked Auto-Encoder
Shitao Xiao | Zheng Liu | Yingxia Shao | Zhao Cao

Despite pre-training’s progress in many important NLP tasks, it remains to explore effective pre-training strategies for dense retrieval. In this paper, we propose RetroMAE, a new retrieval oriented pre-training paradigm based on Masked Auto-Encoder (MAE). RetroMAE is highlighted by three critical designs. 1) A novel MAE workflow, where the input sentence is polluted for encoder and decoder with different masks. The sentence embedding is generated from the encoder’s masked input; then, the original sentence is recovered based on the sentence embedding and the decoder’s masked input via masked language modeling. 2) Asymmetric model structure, with a full-scale BERT like transformer as encoder, and a one-layer transformer as decoder. 3) Asymmetric masking ratios, with a moderate ratio for encoder: 15 30%, and an aggressive ratio for decoder: 50 70%. Our framework is simple to realize and empirically competitive: the pre-trained models dramatically improve the SOTA performances on a wide range of dense retrieval benchmarks, like BEIR and MS MARCO. The source code and pre-trained models are made publicly available at so as to inspire more interesting research.

Aligning Recommendation and Conversation via Dual Imitation
Jinfeng Zhou | Bo Wang | Minlie Huang | Dongming Zhao | Kun Huang | Ruifang He | Yuexian Hou

Human conversations of recommendation naturally involve the shift of interests which can align the recommendation actions and conversation process to make accurate recommendations with rich explanations. However, existing conversational recommendation systems (CRS) ignore the advantage of user interest shift in connecting recommendation and conversation, which leads to an ineffective loose coupling structure of CRS. To address this issue, by modeling the recommendation actions as recommendation paths in a knowledge graph (KG), we propose DICR (Dual Imitation for Conversational Recommendation), which designs a dual imitation to explicitly align the recommendation paths and user interest shift paths in a recommendation module and a conversation module, respectively. By exchanging alignment signals, DICR achieves bidirectional promotion between recommendation and conversation modules and generates high-quality responses with accurate recommendations and coherent explanations. Experiments demonstrate that DICR outperforms the state-of-the-art models on recommendation and conversation performance with automatic, human, and novel explainability metrics.

QRelScore: Better Evaluating Generated Questions with Deeper Understanding of Context-aware Relevance
Xiaoqiang Wang | Bang Liu | Siliang Tang | Lingfei Wu

Existing metrics for assessing question generation not only require costly human reference but also fail to take into account the input context of generation, rendering the lack of deep understanding of the relevance between the generated questions and input contexts. As a result, they may wrongly penalize a legitimate and reasonable candidate question when it (1) involves complicated reasoning with the context or (2) can be grounded by multiple evidences in the context.In this paper, we propose QRelScore, a context-aware Relevance evaluation metric for Question Generation.Based on off-the-shelf language models such as BERT and GPT2, QRelScore employs both word-level hierarchical matching and sentence-level prompt-based generation to cope with the complicated reasoning and diverse generation from multiple evidences, respectively.Compared with existing metrics, our experiments demonstrate that QRelScore is able to achieve a higher correlation with human judgments while being much more robust to adversarial samples.

Abstract Visual Reasoning with Tangram Shapes
Anya Ji | Noriyuki Kojima | Noah Rush | Alane Suhr | Wai Keen Vong | Robert Hawkins | Yoav Artzi

We introduce KiloGram, a resource for studying abstract visual reasoning in humans and machines. Drawing on the history of tangram puzzles as stimuli in cognitive science, we build a richly annotated dataset that, with >1k distinct stimuli, is orders of magnitude larger and more diverse than prior resources. It is both visually and linguistically richer, moving beyond whole shape descriptions to include segmentation maps and part labels. We use this resource to evaluate the abstract visual reasoning capacities of recent multi-modal models. We observe that pre-trained weights demonstrate limited abstract reasoning, which dramatically improves with fine-tuning. We also observe that explicitly describing parts aids abstract reasoning for both humans and models, especially when jointly encoding the linguistic and visual inputs.

UnifiedSKG: Unifying and Multi-Tasking Structured Knowledge Grounding with Text-to-Text Language Models
Tianbao Xie | Chen Henry Wu | Peng Shi | Ruiqi Zhong | Torsten Scholak | Michihiro Yasunaga | Chien-Sheng Wu | Ming Zhong | Pengcheng Yin | Sida I. Wang | Victor Zhong | Bailin Wang | Chengzu Li | Connor Boyle | Ansong Ni | Ziyu Yao | Dragomir Radev | Caiming Xiong | Lingpeng Kong | Rui Zhang | Noah A. Smith | Luke Zettlemoyer | Tao Yu

Structured knowledge grounding (SKG) leverages structured knowledge to complete user requests, such as semantic parsing over databases and question answering over knowledge bases. Since the inputs and outputs of SKG tasks are heterogeneous, they have been studied separately by different communities, which limits systematic and compatible research on SKG. In this paper, we overcome this limitation by proposing the UnifiedSKG framework, which unifies 21 SKG tasks into a text-to-text format, aiming to promote systematic SKG research, instead of being exclusive to a single task, domain, or dataset. We use UnifiedSKG to benchmark T5 with different sizes and show that T5, with simple modifications when necessary, achieves state-of-the-art performance on almost all of the 21 tasks. We further demonstrate that multi-task prefix-tuning improves the performance on most tasks, largely improving the overall performance. UnifiedSKG also facilitates the investigation of zero-shot and few-shot learning, and we show that T0, GPT-3, and Codex struggle in zero-shot and few-shot learning for SKG. We also use UnifiedSKG to conduct a series of controlled experiments on structured knowledge encoding variants across SKG tasks. UnifiedSKG is easily extensible to more tasks, and it is open-sourced at

Balanced Adversarial Training: Balancing Tradeoffs between Fickleness and Obstinacy in NLP Models
Hannah Chen | Yangfeng Ji | David Evans

Traditional (fickle) adversarial examples involve finding a small perturbation that does not change an input’s true label but confuses the classifier into outputting a different prediction. Conversely, obstinate adversarial examples occur when an adversary finds a small perturbation that preserves the classifier’s prediction but changes the true label of an input.Adversarial training and certified robust training have shown some effectiveness in improving the robustness of machine learnt models to fickle adversarial examples. We show that standard adversarial training methods focused on reducing vulnerability to fickle adversarial examples may make a model more vulnerable to obstinate adversarial examples, with experiments for both natural language inference and paraphrase identification tasks. To counter this phenomenon, we introduce Balanced Adversarial Training, which incorporates contrastive learning to increase robustness against both fickle and obstinate adversarial examples.

When Can Transformers Ground and Compose: Insights from Compositional Generalization Benchmarks
Ankur Sikarwar | Arkil Patel | Navin Goyal

Humans can reason compositionally whilst grounding language utterances to the real world. Recent benchmarks like ReaSCAN (Wu et al., 2021) use navigation tasks grounded in a grid world to assess whether neural models exhibit similar capabilities. In this work, we present a simple transformer-based model that outperforms specialized architectures on ReaSCAN and a modified version (Qiu et al., 2021) of gSCAN (Ruis et al., 2020). On analyzing the task, we find that identifying the target location in the grid world is the main challenge for the models. Furthermore, we show that a particular split in ReaSCAN, which tests depth generalization, is unfair. On an amended version of this split, we show that transformers can generalize to deeper input structures. Finally, we design a simpler grounded compositional generalization task, RefEx, to investigate how transformers reason compositionally. We show that a single self-attention layer with a single head generalizes to novel combinations of object attributes. Moreover, we derive a precise mathematical construction of the transformer’s computations from the learned network. Overall, we provide valuable insights about the grounded compositional generalization task and the behaviour of transformers on it, which would be useful for researchers working in this area.

Generative Language Models for Paragraph-Level Question Generation
Asahi Ushio | Fernando Alva-Manchego | Jose Camacho-Collados

Powerful generative models have led to recent progress in question generation (QG). However, it is difficult to measure advances in QG research since there are no standardized resources that allow a uniform comparison among approaches. In this paper, we introduce QG-Bench, a multilingual and multidomain benchmark for QG that unifies existing question answering datasets by converting them to a standard QG setting. It includes general-purpose datasets such as SQuAD for English, datasets from ten domains and two styles, as well as datasets in eight different languages. Using QG-Bench as a reference, we perform an extensive analysis of the capabilities of language models for the task. First, we propose robust QG baselines based on fine-tuning generative language models. Then, we complement automatic evaluation based on standard metrics with an extensive manual evaluation, which in turn sheds light on the difficulty of evaluating QG models. Finally, we analyse both the domain adaptability of these models as well as the effectiveness of multilingual models in languages other than English.QG-Bench is released along with the fine-tuned models presented in the paper (, which are also available as a demo (

A Unified Encoder-Decoder Framework with Entity Memory
Zhihan Zhang | Wenhao Yu | Chenguang Zhu | Meng Jiang

Entities, as important carriers of real-world knowledge, play a key role in many NLP tasks.We focus on incorporating entity knowledge into an encoder-decoder framework for informative text generation. Existing approaches tried to index, retrieve, and read external documents as evidence, but they suffered from a large computational overhead. In this work, we propose an encoder-decoder framework with an entity memory, namely EDMem. The entity knowledge is stored in the memory as latent representations, and the memory is pre-trained on Wikipedia along with encoder-decoder parameters. To precisely generate entity names, we design three decoding methods to constrain entity generation by linking entities in the memory. EDMem is a unified framework that can be used on various entity-intensive question answering and generation tasks. Extensive experimental results show that EDMem outperforms both memory-based auto-encoder models and non-memory encoder-decoder models.

Segmenting Numerical Substitution Ciphers
Nada Aldarrab | Jonathan May

Deciphering historical substitution ciphers is a challenging problem. Example problems that have been previously studied include detecting cipher type, detecting plaintext language, and acquiring the substitution key for segmented ciphers. However, attacking unsegmented ciphers is still a challenging task. Segmentation (i.e. finding substitution units) is essential for cracking those ciphers. In this work, we propose the first automatic methods to segment those ciphers using Byte Pair Encoding (BPE) and unigram language models. Our methods achieve an average segmentation error of 2% on 100 randomly-generated monoalphabetic ciphers and 27% on 3 real historical homophonic ciphers. We also propose a method for solving non-deterministic ciphers with existing keys using a lattice and a pretrained language model. Our method leads to the full solution of the IA cipher; a real historical cipher that has not been fully solved until this work.

Crossmodal-3600: A Massively Multilingual Multimodal Evaluation Dataset
Ashish V. Thapliyal | Jordi Pont Tuset | Xi Chen | Radu Soricut

Research in massively multilingual image captioning has been severely hampered by a lack of high-quality evaluation datasets. In this paper we present the Crossmodal-3600 dataset (XM3600 in short), a geographically diverse set of 3600 images annotated with human-generated reference captions in 36 languages. The images were selected from across the world, covering regions where the 36 languages are spoken, and annotated with captions that achieve consistency in terms of style across all languages, while avoiding annotation artifacts due to direct translation. We apply this benchmark to model selection for massively multilingual image captioning models, and show superior correlation results with human evaluations when using XM3600 as golden references for automatic metrics.

ReSel: N-ary Relation Extraction from Scientific Text and Tables by Learning to Retrieve and Select
Yuchen Zhuang | Yinghao Li | Junyang Zhang | Yue Yu | Yingjun Mou | Xiang Chen | Le Song | Chao Zhang

We study the problem of extracting N-ary relation tuples from scientific articles. This task is challenging because the target knowledge tuples can reside in multiple parts and modalities of the document. Our proposed method ReSel decomposes this task into a two-stage procedure that first retrieves the most relevant paragraph/table and then selects the target entity from the retrieved component. For the high-level retrieval stage, ReSel designs a simple and effective feature set, which captures multi-level lexical and semantic similarities between the query and components. For the low-level selection stage, ReSel designs a cross-modal entity correlation graph along with a multi-view architecture, which models both semantic and document-structural relations between entities. Our experiments on three scientific information extraction datasets show that ReSel outperforms state-of-the-art baselines significantly.

GammaE: Gamma Embeddings for Logical Queries on Knowledge Graphs
Dong Yang | Peijun Qing | Yang Li | Haonan Lu | Xiaodong Lin

Embedding knowledge graphs (KGs) for multi-hop logical reasoning is a challenging problem due to massive and complicated structures in many KGs. Recently, many promising works projected entities and queries into a geometric space to efficiently find answers. However, it remains challenging to model the negation and union operator. The negation operator has no strict boundaries, which generates overlapped embeddings and leads to obtaining ambiguous answers. An additional limitation is that the union operator is non-closure, which undermines the model to handle a series of union operators. To address these problems, we propose a novel probabilistic embedding model, namely Gamma Embeddings (GammaE), for encoding entities and queries to answer different types of FOL queries on KGs. We utilize the linear property and strong boundary support of the Gamma distribution to capture more features of entities and queries, which dramatically reduces model uncertainty. Furthermore, GammaE implements the Gamma mixture method to design the closed union operator. The performance of GammaE is validated on three large logical query datasets. Experimental results show that GammaE significantly outperforms state-of-the-art models on public benchmarks.

Reasoning Like Program Executors
Xinyu Pi | Qian Liu | Bei Chen | Morteza Ziyadi | Zeqi Lin | Qiang Fu | Yan Gao | Jian-Guang Lou | Weizhu Chen

Reasoning over natural language is a long-standing goal for the research community. However, studies have shown that existing language models are inadequate in reasoning. To address the issue, we present POET, a novel reasoning pre-training paradigm. Through pre-training language models with programs and their execution results, POET empowers language models to harvest the reasoning knowledge possessed by program executors via a data-driven approach. POET is conceptually simple and can be instantiated by different kinds of program executors. In this paper, we showcase two simple instances POET-Math and POET-Logic, in addition to a complex instance, POET-SQL. Experimental results on six benchmarks demonstrate that POET can significantly boost model performance in natural language reasoning, such as numerical reasoning, logical reasoning, and multi-hop reasoning. POET opens a new gate on reasoning-enhancement pre-training, and we hope our analysis would shed light on the future research of reasoning like program executors.

SEM-F1: an Automatic Way for Semantic Evaluation of Multi-Narrative Overlap Summaries at Scale
Naman Bansal | Mousumi Akter | Shubhra Kanti Karmaker Santu

Recent work has introduced an important yet relatively under-explored NLP task called Semantic Overlap Summarization (SOS) that entails generating a summary from multiple alternative narratives which conveys the common information provided by those narratives. Previous work also published a benchmark dataset for this task by collecting 2,925 alternative narrative pairs from the web and manually annotating 411 different reference summaries by engaging human annotators. In this paper, we exclusively focus on the automated evaluation of the SOS task using the benchmark dataset. More specifically, we first use the popular ROUGE metric from text-summarization literature and conduct a systematic study to evaluate the SOS task. Our experiments discover that ROUGE is not suitable for this novel task and therefore, we propose a new sentence-level precision-recall style automated evaluation metric, called SEM-F1 (Semantic F1). It is inspired by the benefits of the sentence-wise annotation technique using overlap labels reported by the previous work. Our experiments show that the proposed SEM-F1 metric yields a higher correlation with human judgment and higher inter-rater agreement compared to the ROUGE metric.

Inducer-tuning: Connecting Prefix-tuning and Adapter-tuning
Yifan Chen | Devamanyu Hazarika | Mahdi Namazifar | Yang Liu | Di Jin | Dilek Hakkani-Tur

Prefix-tuning, or more generally continuous prompt tuning, has become an essential paradigm of parameter-efficient transfer learning. Using a large pre-trained language model (PLM), prefix-tuning can obtain strong performance by training only a small portion of parameters. In this paper, we propose to understand and further develop prefix-tuning through the kernel lens. Specifically, we make an analogy between prefixes and inducing variables in kernel methods and hypothesize that prefixes serving as inducing variables would improve their overall mechanism. From the kernel estimator perspective, we suggest a new variant of prefix-tuning—inducer-tuning, which shares the exact mechanism as prefix-tuning while leveraging the residual form found in adapter-tuning. This mitigates the initialization issue in prefix-tuning. Through comprehensive empirical experiments on natural language understanding and generation tasks, we demonstrate that inducer-tuning can close the performance gap between prefix-tuning and fine-tuning.

DocInfer: Document-level Natural Language Inference using Optimal Evidence Selection
Puneet Mathur | Gautam Kunapuli | Riyaz Bhat | Manish Shrivastava | Dinesh Manocha | Maneesh Singh

We present DocInfer - a novel, end-to-end Document-level Natural Language Inference model that builds a hierarchical document graph enriched through inter-sentence relations (topical, entity-based, concept-based), performs paragraph pruning using the novel SubGraph Pooling layer, followed by optimal evidence selection based on REINFORCE algorithm to identify the most important context sentences for a given hypothesis. Our evidence selection mechanism allows it to transcend the input length limitation of modern BERT-like Transformer models while presenting the entire evidence together for inferential reasoning. We show this is an important property needed to reason on large documents where the evidence may be fragmented and located arbitrarily far from each other. Extensive experiments on popular corpora - DocNLI, ContractNLI, and ConTRoL datasets, and our new proposed dataset called CaseHoldNLI on the task of legal judicial reasoning, demonstrate significant performance gains of 8-12% over SOTA methods. Our ablation studies validate the impact of our model. Performance improvement of 3-6% on annotation-scarce downstream tasks of fact verification, multiple-choice QA, and contract clause retrieval demonstrates the usefulness of DocInfer beyond primary NLI tasks.

LightEA: A Scalable, Robust, and Interpretable Entity Alignment Framework via Three-view Label Propagation
Xin Mao | Wenting Wang | Yuanbin Wu | Man Lan

Entity Alignment (EA) aims to find equivalent entity pairs between KGs, which is the core step to bridging and integrating multi-source KGs. In this paper, we argue that existing complex EA methods inevitably inherit the inborn defects from their neural network lineage: poor interpretability and weak scalability. Inspired by recent studies, we reinvent the classical Label Propagation algorithm to effectively run on KGs and propose a neural-free EA framework — LightEA, consisting of three efficient components: (i) Random Orthogonal Label Generation, (ii) Three-view Label Propagation, and (iii) Sparse Sinkhorn Operation.According to the extensive experiments on public datasets, LightEA has impressive scalability, robustness, and interpretability. With a mere tenth of time consumption, LightEA achieves comparable results to state-of-the-art methods across all datasets and even surpasses them on many. Besides, due to the computational process of LightEA being entirely linear, we could trace the propagation process at each step and clearly explain how the entities are aligned.

Metric-guided Distillation: Distilling Knowledge from the Metric to Ranker and Retriever for Generative Commonsense Reasoning
Xingwei He | Yeyun Gong | A-Long Jin | Weizhen Qi | Hang Zhang | Jian Jiao | Bartuer Zhou | Biao Cheng | Sm Yiu | Nan Duan

Commonsense generation aims to generate a realistic sentence describing a daily scene under the given concepts, which is very challenging, since it requires models to have relational reasoning and compositional generalization capabilities. Previous work focuses on retrieving prototype sentences for the provided concepts to assist generation. They first use a sparse retriever to retrieve candidate sentences, then re-rank the candidates with a ranker. However, the candidates returned by their ranker may not be the most relevant sentences, since the ranker treats all candidates equally without considering their relevance to the reference sentences of the given concepts. Another problem is that re-ranking is very expensive, but only using retrievers will seriously degrade the performance of their generation models. To solve these problems, we propose the metric distillation rule to distill knowledge from the metric (e.g., BLEU) to the ranker. We further transfer the critical knowledge summarized by the distilled ranker to the retriever. In this way, the relevance scores of candidate sentences predicted by the ranker and retriever will be more consistent with their quality measured by the metric. Experimental results on the CommonGen benchmark verify the effectiveness of our proposed method: (1) Our generation model with the distilled ranker achieves a new state-of-the-art result. (2) Our generation model with the distilled retriever even surpasses the previous SOTA.

Efficient Document Retrieval by End-to-End Refining and Quantizing BERT Embedding with Contrastive Product Quantization
Zexuan Qiu | Qinliang Su | Jianxing Yu | Shijing Si

Efficient document retrieval heavily relies on the technique of semantic hashing, which learns a binary code for every document and employs Hamming distance to evaluate document distances. However, existing semantic hashing methods are mostly established on outdated TFIDF features, which obviously do not contain lots of important semantic information about documents. Furthermore, the Hamming distance can only be equal to one of several integer values, significantly limiting its representational ability for document distances. To address these issues, in this paper, we propose to leverage BERT embeddings to perform efficient retrieval based on the product quantization technique, which will assign for every document a real-valued codeword from the codebook, instead of a binary code as in semantic hashing. Specifically, we first transform the original BERT embeddings via a learnable mapping and feed the transformed embedding into a probabilistic product quantization module to output the assigned codeword. The refining and quantizing modules can be optimized in an end-to-end manner by minimizing the probabilistic contrastive loss. A mutual information maximization based method is further proposed to improve the representativeness of codewords, so that documents can be quantized more accurately. Extensive experiments conducted on three benchmarks demonstrate that our proposed method significantly outperforms current state-of-the-art baselines.

Curriculum Knowledge Distillation for Emoji-supervised Cross-lingual Sentiment Analysis
Jianyang Zhang | Tao Liang | Mingyang Wan | Guowu Yang | Fengmao Lv

Existing sentiment analysis models have achieved great advances with the help of sufficient sentiment annotations. Unfortunately, many languages do not have sufficient sentiment corpus. To this end, recent studies have proposed cross-lingual sentiment analysis to transfer sentiment analysis models from resource-rich languages to low-resource languages. However, these studies either rely on external cross-lingual supervision (e.g., parallel corpora and translation model), or are limited by the cross-lingual gaps. In this work, based on the intuitive assumption that the relationships between emojis and sentiments are consistent across different languages, we investigate transferring sentiment knowledge across languages with the help of emojis. To this end, we propose a novel cross-lingual sentiment analysis approach dubbed Curriculum Knowledge Distiller (CKD). The core idea of CKD is to use emojis to bridge the source and target languages. Note that, compared with texts, emojis are more transferable, but cannot reveal the precise sentiment. Thus, we distill multiple Intermediate Sentiment Classifiers (ISC) on source language corpus with emojis to get ISCs with different attention weights of texts. To transfer them into the target language, we distill ISCs into the Target Language Sentiment Classifier (TSC) following the curriculum learning mechanism. In this way, TSC can learn delicate sentiment knowledge, meanwhile, avoid being affected by cross-lingual gaps. Experimental results on five cross-lingual benchmarks clearly verify the effectiveness of our approach.

Correctable-DST: Mitigating Historical Context Mismatch between Training and Inference for Improved Dialogue State Tracking
Hongyan Xie | Haoxiang Su | Shuangyong Song | Hao Huang | Bo Zou | Kun Deng | Jianghua Lin | Zhihui Zhang | Xiaodong He

Recently proposed dialogue state tracking (DST) approaches predict the dialogue state of a target turn sequentially based on the previous dialogue state. During the training time, the ground-truth previous dialogue state is utilized as the historical context. However, only the previously predicted dialogue state can be used in inference. This discrepancy might lead to error propagation, i.e., mistakes made by the model in the current turn are likely to be carried over to the following turns.To solve this problem, we propose Correctable Dialogue State Tracking (Correctable-DST). Specifically, it consists of three stages: (1) a Predictive State Simulator is exploited to generate a previously “predicted” dialogue state based on the ground-truth previous dialogue state during training; (2) a Slot Detector is proposed to determine the slots with an incorrect value in the previously “predicted” state and the slots whose values are to be updated in the current turn; (3) a State Generator takes the name of the above-selected slots as a prompt to generate the current state.Empirical results show that our approach achieves 67.51%, 68.24%, 70.30%, 71.38%, and 81.27% joint goal accuracy on MultiWOZ 2.0-2.4 datasets, respectively, and achieves a new state-of-the-art performance with significant improvements.

DropMix: A Textual Data Augmentation Combining Dropout with Mixup
Fanshuang Kong | Richong Zhang | Xiaohui Guo | Samuel Mensah | Yongyi Mao

Overfitting is a notorious problem when there is insufficient data to train deep neural networks in machine learning tasks. Data augmentation regularization methods such as Dropout, Mixup, and their enhanced variants are effective and prevalent, and achieve promising performance to overcome overfitting. However, in text learning, most of the existing regularization approaches merely adopt ideas from computer vision without considering the importance of dimensionality in natural language processing. In this paper, we argue that the property is essential to overcome overfitting in text learning. Accordingly, we present a saliency map informed textual data augmentation and regularization framework, which combines Dropout and Mixup, namely DropMix, to mitigate the overfitting problem in text learning. In addition, we design a procedure that drops and patches fine grained shapes of the saliency map under the DropMix framework to enhance regularization. Empirical studies confirm the effectiveness of the proposed approach on 12 text classification tasks.

Cross-document Event Coreference Search: Task, Dataset and Modeling
Alon Eirew | Avi Caciularu | Ido Dagan

The task of Cross-document Coreference Resolution has been traditionally formulated as requiring to identify all coreference links across a given set of documents. We propose an appealing, and often more applicable, complementary set up for the task – Cross-document Coreference Search, focusing in this paper on event coreference. Concretely, given a mention in context of an event of interest, considered as a query, the task is to find all coreferring mentions for the query event in a large document collection. To support research on this task, we create a corresponding dataset, which is derived from Wikipedia while leveraging annotations in the available Wikipedia Event Coreferecene dataset (WEC-Eng). Observing that the coreference search setup is largely analogous to the setting of Open Domain Question Answering, we adapt the prominent Deep Passage Retrieval (DPR) model to our setting, as an appealing baseline. Finally, we present a novel model that integrates a powerful coreference scoring scheme into the DPR architecture, yielding improved performance.

VIRT: Improving Representation-based Text Matching via Virtual Interaction
Dan Li | Yang Yang | Hongyin Tang | Jiahao Liu | Qifan Wang | Jingang Wang | Tong Xu | Wei Wu | Enhong Chen

Text matching is a fundamental research problem in natural language understanding. Interaction-based approaches treat the text pair as a single sequence and encode it through cross encoders, while representation-based models encode the text pair independently with siamese or dual encoders. Interaction-based models require dense computations and thus are impractical in real-world applications. Representation-based models have become the mainstream paradigm for efficient text matching. However, these models suffer from severe performance degradation due to the lack of interactions between the pair of texts. To remedy this, we propose a Virtual InteRacTion mechanism (VIRT) for improving representation-based text matching while maintaining its efficiency. In particular, we introduce an interactive knowledge distillation module that is only applied during training. It enables deep interaction between texts by effectively transferring knowledge from the interaction-based model. A light interaction strategy is designed to fully leverage the learned interactive knowledge. Experimental results on six text matching benchmarks demonstrate the superior performance of our method over several state-of-the-art representation-based models. We further show that VIRT can be integrated into existing methods as plugins to lift their performances.

MAVEN-ERE: A Unified Large-scale Dataset for Event Coreference, Temporal, Causal, and Subevent Relation Extraction
Xiaozhi Wang | Yulin Chen | Ning Ding | Hao Peng | Zimu Wang | Yankai Lin | Xu Han | Lei Hou | Juanzi Li | Zhiyuan Liu | Peng Li | Jie Zhou

The diverse relationships among real-world events, including coreference, temporal, causal, and subevent relations, are fundamental to understanding natural languages. However, two drawbacks of existing datasets limit event relation extraction (ERE) tasks: (1) Small scale. Due to the annotation complexity, the data scale of existing datasets is limited, which cannot well train and evaluate data-hungry models. (2) Absence of unified annotation. Different types of event relations naturally interact with each other, but existing datasets only cover limited relation types at once, which prevents models from taking full advantage of relation interactions. To address these issues, we construct a unified large-scale human-annotated ERE dataset MAVEN-ERE with improved annotation schemes. It contains 103,193 event coreference chains, 1,216,217 temporal relations, 57,992 causal relations, and 15,841 subevent relations, which is larger than existing datasets of all the ERE tasks by at least an order of magnitude. Experiments show that ERE on MAVEN-ERE is quite challenging, and considering relation interactions with joint learning can improve performances. The dataset and source codes can be obtained from

Entity Extraction in Low Resource Domains with Selective Pre-training of Large Language Models
Aniruddha Mahapatra | Sharmila Reddy Nangi | Aparna Garimella | Anandhavelu N

Transformer-based language models trained on large natural language corpora have been very useful in downstream entity extraction tasks. However, they often result in poor performances when applied to domains that are different from those they are pretrained on. Continued pretraining using unlabeled data from target domains can help improve the performances of these language models on the downstream tasks. However, using all of the available unlabeled data for pretraining can be time-intensive; also, it can be detrimental to the performance of the downstream tasks, if the unlabeled data is not aligned with the data distribution for the target tasks. Previous works employed external supervision in the form of ontologies for selecting appropriate data samples for pretraining, but external supervision can be quite hard to obtain in low-resource domains. In this paper, we introduce effective ways to select data from unlabeled corpora of target domains for language model pretraining to improve the performances in target entity extraction tasks. Our data selection strategies do not require any external supervision. We conduct extensive experiments for the task of named entity recognition (NER) on seven different domains and show that language models pretrained on target domain unlabeled data obtained using our data selection strategies achieve better performances compared to those using data selection strategies in previous works that use external supervision. We also show that these pretrained language models using our data selection strategies outperform those pretrained on all of the available unlabeled target domain data.

How Large Language Models are Transforming Machine-Paraphrase Plagiarism
Jan Philip Wahle | Terry Ruas | Frederic Kirstein | Bela Gipp

The recent success of large language models for text generation poses a severe threat to academic integrity, as plagiarists can generate realistic paraphrases indistinguishable from original work.However, the role of large autoregressive models in generating machine-paraphrased plagiarism and their detection is still incipient in the literature.This work explores T5 and GPT3 for machine-paraphrase generation on scientific articles from arXiv, student theses, and Wikipedia.We evaluate the detection performance of six automated solutions and one commercial plagiarism detection software and perform a human study with 105 participants regarding their detection performance and the quality of generated examples.Our results suggest that large language models can rewrite text humans have difficulty identifying as machine-paraphrased (53% mean acc.).Human experts rate the quality of paraphrases generated by GPT-3 as high as original texts (clarity 4.0/5, fluency 4.2/5, coherence 3.8/5).The best-performing detection model (GPT-3) achieves 66% F1-score in detecting paraphrases.We make our code, data, and findings publicly available to facilitate the development of detection solutions.

M2D2: A Massively Multi-Domain Language Modeling Dataset
Machel Reid | Victor Zhong | Suchin Gururangan | Luke Zettlemoyer

We present M2D2, a fine-grained, massively multi-domain corpus for studying domain adaptation in language models (LMs). M2D2 consists of 8.5B tokens and spans 145 domains extracted from Wikipedia and Semantic Scholar. Using ontologies derived from Wikipedia and ArXiv categories, we organize the domains in each data source into 22 groups. This two-level hierarchy enables the study of relationships between domains and their effects on in- and out-of-domain performance after adaptation. We also present a number of insights into the nature of effective domain adaptation in LMs, as examples of the new types of studies M2D2 enables. To improve in-domain performance, we show the benefits of adapting the LM along a domain hierarchy; adapting to smaller amounts of fine-grained domain-specific data can lead to larger in-domain performance gains than larger amounts of weakly relevant data. We further demonstrate a trade-off between in-domain specialization and out-of-domain generalization within and across ontologies, as well as a strong correlation between out-of-domain performance and lexical overlap between domains.

“Will You Find These Shortcuts?” A Protocol for Evaluating the Faithfulness of Input Salience Methods for Text Classification
Jasmijn Bastings | Sebastian Ebert | Polina Zablotskaia | Anders Sandholm | Katja Filippova

Feature attribution a.k.a. input salience methods which assign an importance score to a feature are abundant but may produce surprisingly different results for the same model on the same input. While differences are expected if disparate definitions of importance are assumed, most methods claim to provide faithful attributions and point at the features most relevant for a model’s prediction. Existing work on faithfulness evaluation is not conclusive and does not provide a clear answer as to how different methods are to be compared.Focusing on text classification and the model debugging scenario, our main contribution is a protocol for faithfulness evaluation that makes use of partially synthetic data to obtain ground truth for feature importance ranking. Following the protocol, we do an in-depth analysis of four standard salience method classes on a range of datasets and lexical shortcuts for BERT and LSTM models. We demonstrate that some of the most popular method configurations provide poor results even for simple shortcuts while a method judged to be too simplistic works remarkably well for BERT.

Information-Transport-based Policy for Simultaneous Translation
Shaolei Zhang | Yang Feng

Simultaneous translation (ST) outputs translation while receiving the source inputs, and hence requires a policy to determine whether to translate a target token or wait for the next source token. The major challenge of ST is that each target token can only be translated based on the current received source tokens, where the received source information will directly affect the translation quality. So naturally, how much source information is received for the translation of the current target token is supposed to be the pivotal evidence for the ST policy to decide between translating and waiting. In this paper, we treat the translation as information transport from source to target and accordingly propose an Information-Transport-based Simultaneous Translation (ITST). ITST quantifies the transported information weight from each source token to the current target token, and then decides whether to translate the target token according to its accumulated received information. Experiments on both text-to-text ST and speech-to-text ST (a.k.a., streaming speech translation) tasks show that ITST outperforms strong baselines and achieves state-of-the-art performance.

Learning to Adapt to Low-Resource Paraphrase Generation
Zhigen Li | Yanmeng Wang | Rizhao Fan | Ye Wang | Jianfeng Li | Shaojun Wang

Paraphrase generation is a longstanding NLP task and achieves great success with the aid of large corpora. However, transferring a paraphrasing model to another domain encounters the problem of domain shifting especially when the data is sparse. At the same time, widely using large pre-trained language models (PLMs) faces the overfitting problem when training on scarce labeled data. To mitigate these two issues, we propose, LAPA, an effective adapter for PLMs optimized by meta-learning. LAPA has three-stage training on three types of related resources to solve this problem: 1. pre-training PLMs on unsupervised corpora, 2. inserting an adapter layer and meta-training on source domain labeled data, and 3. fine-tuning adapters on a small amount of target domain labeled data. This method enables paraphrase generation models to learn basic language knowledge first, then learn the paraphrasing task itself later, and finally adapt to the target task. Our experimental results demonstrate that LAPA achieves state-of-the-art in supervised, unsupervised, and low-resource settings on three benchmark datasets. With only 2% of trainable parameters and 1% labeled data of the target task, our approach can achieve a competitive performance with previous work.

A Distributional Lens for Multi-Aspect Controllable Text Generation
Yuxuan Gu | Xiaocheng Feng | Sicheng Ma | Lingyuan Zhang | Heng Gong | Bing Qin

Multi-aspect controllable text generation is a more challenging and practical task than single-aspect control. Existing methods achieve complex multi-aspect control by fusing multiple controllers learned from single-aspect, but suffer from attribute degeneration caused by the mutual interference of these controllers. To address this, we provide observations on attribute fusion from a distributional perspective and propose to directly search for the intersection areas of multiple attribute distributions as their combination for generation. Our method first estimates the attribute space with an autoencoder structure. Afterward, we iteratively approach the intersections by jointly minimizing distances to points representing different attributes. Finally, we map them to attribute-relevant sentences with a prefix-tuning-based decoder. Experiments on the three-aspect control task, including sentiment, topic, and detoxification aspects, reveal that our method outperforms several strong baselines on attribute relevance and text quality and achieves the SOTA. Further analysis also supplies some explanatory support for the effectiveness of our approach.

ELMER: A Non-Autoregressive Pre-trained Language Model for Efficient and Effective Text Generation
Junyi Li | Tianyi Tang | Wayne Xin Zhao | Jian-Yun Nie | Ji-Rong Wen

We study the text generation task under the approach of pre-trained language models (PLMs). Typically, an auto-regressive (AR) method is adopted for generating texts in a token-by-token manner. Despite many advantages of AR generation, it usually suffers from inefficient inference. Therefore, non-autoregressive (NAR) models are proposed to generate all target tokens simultaneously. However, NAR models usually generate texts of lower quality due to the absence of token dependency in the output text. In this paper, we propose ELMER: an efficient and effective PLM for NAR text generation to explicitly model the token dependency during NAR generation. By leveraging the early exit technique, ELMER enables the token generations at different layers, according to their prediction confidence (a more confident token will exit at a lower layer). Besides, we propose a novel pre-training objective, Layer Permutation Language Modeling, to pre-train ELMER by permuting the exit layer for each token in sequences. Experiments on three text generation tasks show that ELMER significantly outperforms NAR models and further narrows the performance gap with AR PLMs (ELMER (29.92) vs BART (30.61) ROUGE-L in XSUM) while achieving over 10 times inference speedup.

Multilingual Relation Classification via Efficient and Effective Prompting
Yuxuan Chen | David Harbecke | Leonhard Hennig

Prompting pre-trained language models has achieved impressive performance on various NLP tasks, especially in low data regimes. Despite the success of prompting in monolingual settings, applying prompt-based methods in multilingual scenarios has been limited to a narrow set of tasks, due to the high cost of handcrafting multilingual prompts. In this paper, we present the first work on prompt-based multilingual relation classification (RC), by introducing an efficient and effective method that constructs prompts from relation triples and involves only minimal translation for the class labels. We evaluate its performance in fully supervised, few-shot and zero-shot scenarios, and analyze its effectiveness across 14 languages, prompt variants, and English-task training in cross-lingual settings. We find that in both fully supervised and few-shot scenarios, our prompt method beats competitive baselines: fine-tuning XLM-R_EM and null prompts. It also outperforms the random baseline by a large margin in zero-shot experiments. Our method requires little in-language knowledge and can be used as a strong baseline for similar multilingual classification tasks.

Topic-Regularized Authorship Representation Learning
Jitkapat Sawatphol | Nonthakit Chaiwong | Can Udomcharoenchaikit | Sarana Nutanong

Authorship attribution is a task that aims to identify the author of a given piece of writing. We aim to develop a generalized solution that can handle a large number of texts from authors and topics unavailable in training data. Previous studies have proposed strategies to address only either unseen authors or unseen topics. Authorship representation learning has been shown to work in open-set environments with a large number of unseen authors but has not been explicitly designed for cross-topic environments at the same time. To handle a large number of unseen authors and topics, we propose Authorship Representation Regularization (ARR), a distillation framework that creates authorship representation with reduced reliance on topic-specific information. To assess the performance of our framework, we also propose a cross-topic-open-set evaluation method. Our proposed method has improved performances in the cross-topic-open set setup over baselines in 4 out of 6 cases.

Fine-grained Contrastive Learning for Relation Extraction
William Hogan | Jiacheng Li | Jingbo Shang

Recent relation extraction (RE) works have shown encouraging improvements by conducting contrastive learning on silver labels generated by distant supervision before fine-tuning on gold labels. Existing methods typically assume all these silver labels are accurate and treat them equally; however, distant supervision is inevitably noisy–some silver labels are more reliable than others. In this paper, we propose fine-grained contrastive learning (FineCL) for RE, which leverages fine-grained information about which silver labels are and are not noisy to improve the quality of learned relationship representations for RE. We first assess the quality of silver labels via a simple and automatic approach we call “learning order denoising,” where we train a language model to learn these relations and record the order of learned training instances. We show that learning order largely corresponds to label accuracy–early-learned silver labels have, on average, more accurate labels than later-learned silver labels. Then, during pre-training, we increase the weights of accurate labels within a novel contrastive learning objective. Experiments on several RE benchmarks show that FineCL makes consistent and significant performance gains over state-of-the-art methods.

Curriculum Prompt Learning with Self-Training for Abstractive Dialogue Summarization
Changqun Li | Linlin Wang | Xin Lin | Gerard de Melo | Liang He

Succinctly summarizing dialogue is a task of growing interest, but inherent challenges, such as insufficient training data and low information density impede our ability to train abstractive models. In this work, we propose a novel curriculum-based prompt learning method with self-training to address these problems. Specifically, prompts are learned using a curriculum learning strategy that gradually increases the degree of prompt perturbation, thereby improving the dialogue understanding and modeling capabilities of our model. Unlabeled dialogue is incorporated by means of self-training so as to reduce the dependency on labeled data. We further investigate topic-aware prompts to better plan for the generation of summaries. Experiments confirm that our model substantially outperforms strong baselines and achieves new state-of-the-art results on the AMI and ICSI datasets. Human evaluations also show the superiority of our model with regard to the summary generation quality.

Zero-Shot Text Classification with Self-Training
Ariel Gera | Alon Halfon | Eyal Shnarch | Yotam Perlitz | Liat Ein-Dor | Noam Slonim

Recent advances in large pretrained language models have increased attention to zero-shot text classification. In particular, models finetuned on natural language inference datasets have been widely adopted as zero-shot classifiers due to their promising results and off-the-shelf availability. However, the fact that such models are unfamiliar with the target task can lead to instability and performance issues. We propose a plug-and-play method to bridge this gap using a simple self-training approach, requiring only the class names along with an unlabeled dataset, and without the need for domain expertise or trial and error. We show that fine-tuning the zero-shot classifier on its most confident predictions leads to significant performance gains across a wide range of text classification tasks, presumably since self-training adapts the zero-shot model to the task at hand.

Deconfounding Legal Judgment Prediction for European Court of Human Rights Cases Towards Better Alignment with Experts
T.y.s.s Santosh | Shanshan Xu | Oana Ichim | Matthias Grabmair

This work demonstrates that Legal Judgement Prediction systems without expert-informed adjustments can be vulnerable to shallow, distracting surface signals that arise from corpus construction, case distribution, and confounding factors. To mitigate this, we use domain expertise to strategically identify statistically predictive but legally irrelevant information. We adopt adversarial training to prevent the system from relying on it. We evaluate our deconfounded models by employing interpretability techniques and comparing to expert annotations. Quantitative experiments and qualitative analysis show that our deconfounded model consistently aligns better with expert rationales than baselines trained for prediction only. We further contribute a set of reference expert annotations to the validation and testing partitions of an existing benchmark dataset of European Court of Human Rights cases.

SQuALITY: Building a Long-Document Summarization Dataset the Hard Way
Alex Wang | Richard Yuanzhe Pang | Angelica Chen | Jason Phang | Samuel R. Bowman

Summarization datasets are often assembled either by scraping naturally occurring public-domain summaries—which are nearly always in difficult-to-work-with technical domains—or by using approximate heuristics to extract them from everyday text—which frequently yields unfaithful summaries. In this work, we turn to a slower but more straightforward approach to developing summarization benchmark data: We hire highly-qualified contractors to read stories and write original summaries from scratch. To amortize reading time, we collect five summaries per document, with the first giving an overview and the subsequent four addressing specific questions. We use this protocol to collect SQuALITY, a dataset of question-focused summaries built on the same public-domain short stories as the multiple-choice dataset QuALITY (Pang et al., 2021). Experiments with state-of-the-art summarization systems show that our dataset is challenging and that existing automatic evaluation metrics are weak indicators of quality.

MetaASSIST: Robust Dialogue State Tracking with Meta Learning
Fanghua Ye | Xi Wang | Jie Huang | Shenghui Li | Samuel Stern | Emine Yilmaz

Existing dialogue datasets contain lots of noise in their state annotations. Such noise can hurt model training and ultimately lead to poor generalization performance. A general framework named ASSIST has recently been proposed to train robust dialogue state tracking (DST) models. It introduces an auxiliary model to generate pseudo labels for the noisy training set. These pseudo labels are combined with vanilla labels by a common fixed weighting parameter to train the primary DST model. Notwithstanding the improvements of ASSIST on DST, tuning the weighting parameter is challenging. Moreover, a single parameter shared by all slots and all instances may be suboptimal. To overcome these limitations, we propose a meta learning-based framework MetaASSIST to adaptively learn the weighting parameter. Specifically, we propose three schemes with varying degrees of flexibility, ranging from slot-wise to both slot-wise and instance-wise, to convert the weighting parameter into learnable functions. These functions are trained in a meta-learning manner by taking the validation set as meta data. Experimental results demonstrate that all three schemes can achieve competitive performance. Most impressively, we achieve a state-of-the-art joint goal accuracy of 80.10% on MultiWOZ 2.4.

Multilingual Machine Translation with Hyper-Adapters
Christos Baziotis | Mikel Artetxe | James Cross | Shruti Bhosale

Multilingual machine translation suffers from negative interference across languages. A common solution is to relax parameter sharing with language-specific modules like adapters. However, adapters of related languages are unable to transfer information, and their total number of parameters becomes prohibitively expensive as the number of languages grows. In this work, we overcome these drawbacks using hyper-adapters – hyper-networks that generate adapters from language and layer embeddings. While past work had poor results when scaling hyper-networks, we propose a rescaling fix that significantly improves convergence and enables training larger hyper-networks. We find that hyper-adapters are more parameter efficient than regular adapters, reaching the same performance with up to 12 times less parameters. When using the same number of parameters and FLOPS, our approach consistently outperforms regular adapters. Also, hyper-adapters converge faster than alternative approaches and scale better than regular dense networks. Our analysis shows that hyper-adapters learn to encode language relatedness, enabling positive transfer across languages.

Z-LaVI: Zero-Shot Language Solver Fueled by Visual Imagination
Yue Yang | Wenlin Yao | Hongming Zhang | Xiaoyang Wang | Dong Yu | Jianshu Chen

Large-scale pretrained language models have made significant advances in solving downstream language understanding tasks. However, they generally suffer from reporting bias, the phenomenon describing the lack of explicit commonsense knowledge in written text, e.g., ”an orange is orange”. To overcome this limitation, we develop a novel approach, Z-LaVI, to endow language models with visual imagination capabilities. Specifically, we leverage two complementary types of ”imaginations”: (i) recalling existing images through retrieval and (ii) synthesizing nonexistent images via text-to-image generation. Jointly exploiting the language inputs and the imagination, a pretrained vision-language model (e.g., CLIP) eventually composes a zero-shot solution to the original language tasks. Notably, fueling language models with imagination can effectively leverage visual knowledge to solve plain language tasks. In consequence, Z-LaVI consistently improves the zero-shot performance of existing language models across a diverse set of language tasks.

Using Commonsense Knowledge to Answer Why-Questions
Yash Kumar Lal | Niket Tandon | Tanvi Aggarwal | Horace Liu | Nathanael Chambers | Raymond Mooney | Niranjan Balasubramanian

Answering questions in narratives about why events happened often requires commonsense knowledge external to the text. What aspects of this knowledge are available in large language models? What aspects can be made accessible via external commonsense resources? We study these questions in the context of answering questions in the TellMeWhy dataset using COMET as a source of relevant commonsense relations. We analyze the effects of model size (T5 and GPT3) along with methods of injecting knowledge (COMET) into these models. Results show that the largest models, as expected, yield substantial improvements over base models. Injecting external knowledge helps models of various sizes, but the amount of improvement decreases with larger model size. We also find that the format in which knowledge is provided is critical, and that smaller models benefit more from larger amounts of knowledge. Finally, we develop an ontology of knowledge types and analyze the relative coverage of the models across these categories.

Affective Idiosyncratic Responses to Music
Sky CH-Wang | Evan Li | Oliver Li | Smaranda Muresan | Zhou Yu

Affective responses to music are highly personal. Despite consensus that idiosyncratic factors play a key role in regulating how listeners emotionally respond to music, precisely measuring the marginal effects of these variables has proved challenging. To address this gap, we develop computational methods to measure affective responses to music from over 403M listener comments on a Chinese social music platform. Building on studies from music psychology in systematic and quasi-causal analyses, we test for musical, lyrical, contextual, demographic, and mental health effects that drive listener affective responses. Finally, motivated by the social phenomenon known as 网抑云 (wǎng-yì-yún), we identify influencing factors of platform user self-disclosures, the social support they receive, and notable differences in discloser user activity.

Successive Prompting for Decomposing Complex Questions
Dheeru Dua | Shivanshu Gupta | Sameer Singh | Matt Gardner

Answering complex questions that require making latent decisions is a challenging task, especially when limited supervision is available. Recent works leverage the capabilities of large language models (LMs) to perform complex question answering in a few-shot setting by demonstrating how to output intermediate rationalizations while solving the complex question in a single pass. We introduce “Successive Prompting” where, we iteratively break down a complex task into a simple task, solve it, and then repeat the process until we get the final solution. Successive prompting decouples the supervision for decomposing complex questions from the supervision for answering simple questions, allowing us to (1) have multiple opportunities to query in-context examples at each reasoning step (2) learn question decomposition separately from question answering, including using synthetic data, and (3) use bespoke (fine-tuned) components for reasoning steps where a large LM does not perform well. The intermediate supervision is typically manually written, which can be expensive to collect. We introduce a way to generate synthetic dataset which can be used to bootstrap model’s ability to decompose and answer intermediate questions. Our best model (with successive prompting) achieves an improvement in F1 of ~5% when compared with a state-of-the-art model with synthetic augmentations and few-shot version of the DROP dataset.

Maieutic Prompting: Logically Consistent Reasoning with Recursive Explanations
Jaehun Jung | Lianhui Qin | Sean Welleck | Faeze Brahman | Chandra Bhagavatula | Ronan Le Bras | Yejin Choi

Pre-trained language models (LMs) struggle with consistent reasoning; recently, prompting LMs to generate explanations that self-guide the inference has emerged as a promising direction to amend this. However, these approaches are fundamentally bounded by the correctness of explanations, which themselves are often noisy and inconsistent. In this work, we develop Maieutic Prompting, which aims to infer a correct answer to a question even from the unreliable generations of LM. Maieutic Prompting induces a tree of explanations abductively (e.g. X is true, because ...) and recursively, then frames the inference as a satisfiability problem over these explanations and their logical relations. We test Maieutic Prompting for true/false QA on three challenging benchmarks that require complex commonsense reasoning. Maieutic Prompting achieves up to 20% better accuracy than state-of-the-art prompting methods, and as a fully unsupervised approach, performs competitively with supervised models. We also show that Maieutic Prompting improves robustness in inference while providing interpretable rationales.

DANLI: Deliberative Agent for Following Natural Language Instructions
Yichi Zhang | Jianing Yang | Jiayi Pan | Shane Storks | Nikhil Devraj | Ziqiao Ma | Keunwoo Yu | Yuwei Bao | Joyce Chai

Recent years have seen an increasing amount of work on embodied AI agents that can perform tasks by following human language instructions. However, most of these agents are reactive, meaning that they simply learn and imitate behaviors encountered in the training data. These reactive agents are insufficient for long-horizon complex tasks. To address this limitation, we propose a neuro-symbolic deliberative agent that, while following language instructions, proactively applies reasoning and planning based on its neural and symbolic representations acquired from past experience (e.g., natural language and egocentric vision). We show that our deliberative agent achieves greater than 70% improvement over reactive baselines on the challenging TEACh benchmark. Moreover, the underlying reasoning and planning processes, together with our modular framework, offer impressive transparency and explainability to the behaviors of the agent. This enables an in-depth understanding of the agent’s capabilities, which shed light on challenges and opportunities for future embodied agents for instruction following. The code is available at

Tracing Semantic Variation in Slang
Zhewei Sun | Yang Xu

The meaning of a slang term can vary in different communities. However, slang semantic variation is not well understood and under-explored in the natural language processing of slang. One existing view argues that slang semantic variation is driven by culture-dependent communicative needs. An alternative view focuses on slang’s social functions suggesting that the desire to foster semantic distinction may have led to the historical emergence of community-specific slang senses. We explore these theories using computational models and test them against historical slang dictionary entries, with a focus on characterizing regularity in the geographical variation of slang usages attested in the US and the UK over the past two centuries. We show that our models are able to predict the regional identity of emerging slang word meanings from historical slang records. We offer empirical evidence that both communicative need and semantic distinction play a role in the variation of slang meaning yet their relative importance fluctuates over the course of history. Our work offers an opportunity for incorporating historical cultural elements into the natural language processing of slang.

Fine-grained Category Discovery under Coarse-grained supervision with Hierarchical Weighted Self-contrastive Learning
Wenbin An | Feng Tian | Ping Chen | Siliang Tang | Qinghua Zheng | QianYing Wang

Novel category discovery aims at adapting models trained on known categories to novel categories. Previous works only focus on the scenario where known and novel categories are of the same granularity.In this paper, we investigate a new practical scenario called Fine-grained Category Discovery under Coarse-grained supervision (FCDC). FCDC aims at discovering fine-grained categories with only coarse-grained labeled data, which can adapt models to categories of different granularity from known ones and reduce significant labeling cost. It is also a challenging task since supervised training on coarse-grained categories tends to focus on inter-class distance (distance between coarse-grained classes) but ignore intra-class distance (distance between fine-grained sub-classes) which is essential for separating fine-grained categories.Considering most current methods cannot transfer knowledge from coarse-grained level to fine-grained level, we propose a hierarchical weighted self-contrastive network by building a novel weighted self-contrastive module and combining it with supervised learning in a hierarchical manner.Extensive experiments on public datasets show both effectiveness and efficiency of our model over compared methods.

PLM-based World Models for Text-based Games
Minsoo Kim | Yeonjoon Jung | Dohyeon Lee | Seung-won Hwang

World models have improved the ability of reinforcement learning agents to operate in a sample efficient manner, by being trained to predict plausible changes in the underlying environment. As the core tasks of world models are future prediction and commonsense understanding, our claim is that pre-trained language models (PLMs) already provide a strong base upon which to build world models. Worldformer is a recently proposed world model for text-based game environments, based only partially on PLM and transformers. Our distinction is to fully leverage PLMs as actionable world models in text-based game environments, by reformulating generation as constrained decoding which decomposes actions into verb templates and objects. We show that our model improves future valid action prediction and graph change prediction. Additionally, we show that our model better reflects commonsense than standard PLM.

Prompt-Based Meta-Learning For Few-shot Text Classification
Haoxing Zhang | Xiaofeng Zhang | Haibo Huang | Lei Yu

Few-shot Text Classification predicts the semantic label of a given text with a handful of supporting instances. Current meta-learning methods have achieved satisfying results in various few-shot situations. Still, they often require a large amount of data to construct many few-shot tasks for meta-training, which is not practical in real-world few-shot scenarios. Prompt-tuning has recently proved to be another effective few-shot learner by bridging the gap between pre-train and downstream tasks. In this work, we closely combine the two promising few-shot learning methodologies in structure and propose a Prompt-Based Meta-Learning (PBML) model to overcome the above meta-learning problem by adding the prompting mechanism. PBML assigns label word learning to base-learners and template learning to meta-learner, respectively. Experimental results show state-of-the-art performance on four text classification datasets under few-shot settings, with higher accuracy and good robustness. We demonstrate through low-resource experiments that our method alleviates the shortcoming that meta-learning requires too much data for meta-training. In the end, we use the visualization to interpret and verify that the meta-learning framework can help the prompting method converge better. We release our code to reproduce our experiments.

How well can Text-to-Image Generative Models understand Ethical Natural Language Interventions?
Hritik Bansal | Da Yin | Masoud Monajatipoor | Kai-Wei Chang

Text-to-image generative models have achieved unprecedented success in generating high-quality images based on natural language descriptions. However, it is shown that these models tend to favor specific social groups when prompted with neutral text descriptions (e.g., ‘a photo of a lawyer’). Following Zhao et al. (2021), we study the effect on the diversity of the generated images when adding ethical intervention that supports equitable judgment (e.g., ‘if all individuals can be a lawyer irrespective of their gender’) in the input prompts. To this end, we introduce an Ethical NaTural Language Interventions in Text-to-Image GENeration (ENTIGEN) benchmark dataset to evaluate the change in image generations conditional on ethical interventions across three social axes – gender, skin color, and culture. Through CLIP-based and human evaluation on minDALL.E, DALL.E-mini and Stable Diffusion, we find that the model generations cover diverse social groups while preserving the image quality. In some cases, the generations would be anti-stereotypical (e.g., models tend to create images with individuals that are perceived as man when fed with prompts about makeup) in the presence of ethical intervention. Preliminary studies indicate that a large change in the model predictions is triggered by certain phrases such as ‘irrespective of gender’ in the context of gender bias in the ethical interventions. We release code and annotated data at

Geographic Citation Gaps in NLP Research
Mukund Rungta | Janvijay Singh | Saif M. Mohammad | Diyi Yang

In a fair world, people have equitable opportunities to education, to conduct scientific research, to publish, and to get credit for their work, regardless of where they live. However, it is common knowledge among researchers that a vast number of papers accepted at top NLP venues come from a handful of western countries and (lately) China; whereas, very few papers from Africa and South America get published. Similar disparities are also believed to exist for paper citation counts. In the spirit of “what we do not measure, we cannot improve”, this work asks a series of questions on the relationship between geographical location and publication success (acceptance in top NLP venues and citation impact). We first created a dataset of 70,000 papers from the ACL Anthology, extracted their meta-information, andgenerated their citation network. We then show that not only are there substantial geographical disparities in paper acceptance and citation but also that these disparities persist even when controlling for a number of variables such as venue of publication and sub-field of NLP. Further, despite some steps taken by the NLP community to improve geographical diversity, we show that the disparity in publication metrics across locations is still on an increasing trend since the early 2000s. We release our code and dataset here:

Language Models of Code are Few-Shot Commonsense Learners
Aman Madaan | Shuyan Zhou | Uri Alon | Yiming Yang | Graham Neubig

We address the general task of structured commonsense reasoning: given a natural language input, the goal is to generate a graph such as an event or a reasoning-graph.To employ large language models (LMs) for this task, existing approaches ‘serialize’ the output graph as a flat list of nodes and edges.Although feasible, these serialized graphs strongly deviate from the natural language corpora that LMs were pre-trained on, hindering LMs from generating them correctly. In this paper, we show that when we instead frame structured commonsense reasoning tasks as code generation tasks, pre-trained LMs of code are better structured commonsense reasoners than LMs of natural language, even when the downstream task does not involve source code at all.We demonstrate our approach across three diverse structured commonsense reasoning tasks. In all these natural language tasks, we show that using our approach, a code generation LM (codex) outperforms natural-LMs that are fine-tuned on the target task (T5) and other strong LMs such as GPT-3 in the few-shot setting.

Numerical Optimizations for Weighted Low-rank Estimation on Language Models
Ting Hua | Yen-Chang Hsu | Felicity Wang | Qian Lou | Yilin Shen | Hongxia Jin

Singular value decomposition (SVD) is one of the most popular compression methods that approximate a target matrix with smaller matrices. However, standard SVD treats the parameters within the matrix with equal importance, which is a simple but unrealistic assumption. The parameters of a trained neural network model may affect the task performance unevenly, which suggests non-equal importance among the parameters. Compared to SVD, the decomposition method aware of parameter importance is the more practical choice in real cases. Unlike standard SVD, weighed value decomposition is a non-convex optimization problem that lacks a closed-form solution. We systematically investigated multiple optimization strategies to tackle the problem and examined our method by compressing Transformer-based language models.Further, we designed a metric to predict when the SVD may introduce a significant performance drop, for which our method can be a rescue strategy.The extensive evaluations demonstrate that our method can perform better than current SOTA methods in compressing Transformer-based language models.

Generative Multi-hop Retrieval
Hyunji Lee | Sohee Yang | Hanseok Oh | Minjoon Seo

A common practice for text retrieval is to use an encoder to map the documents and the query to a common vector space and perform a nearest neighbor search (NNS); multi-hop retrieval also often adopts the same paradigm, usually with a modification of iteratively reformulating the query vector so that it can retrieve different documents at each hop. However, such a bi-encoder approach has limitations in multi-hop settings; (1) the reformulated query gets longer as the number of hops increases, which further tightens the embedding bottleneck of the query vector, and (2) it is prone to error propagation. In this paper, we focus on alleviating these limitations in multi-hop settings by formulating the problem in a fully generative way. We propose an encoder-decoder model that performs multi-hop retrieval by simply generating the entire text sequences of the retrieval targets, which means the query and the documents interact in the language model’s parametric space rather than L2 or inner product space as in the bi-encoder approach. Our approach, Generative Multi-hop Retrieval (GMR), consistently achieves comparable or higher performance than bi-encoder models in five datasets while demonstrating superior GPU memory and storage footprint.

Visual Spatial Description: Controlled Spatial-Oriented Image-to-Text Generation
Yu Zhao | Jianguo Wei | ZhiChao Lin | Yueheng Sun | Meishan Zhang | Min Zhang

Image-to-text tasks such as open-ended image captioning and controllable image description have received extensive attention for decades. Here we advance this line of work further, presenting Visual Spatial Description (VSD), a new perspective for image-to-text toward spatial semantics. Given an image and two objects inside it, VSD aims to produce one description focusing on the spatial perspective between the two objects. Accordingly, we annotate a dataset manually to facilitate the investigation of the newly-introduced task, and then build several benchmark encoder-decoder models by using VL-BART and VL-T5 as backbones. In addition, we investigate visual spatial relationship classification (VSRC) information into our model by pipeline and end-to-end architectures. Finally, we conduct experiments on our benchmark dataset to evaluate all our models. Results show that our models are awe-inspiring, offering accurate and human-like spatial-oriented text descriptions. Besides, VSRC has great potential for VSD, and the joint end-to-end architecture is the better choice for their integration. We will make the dataset and codes publicly available for research purposes.

M3: A Multi-View Fusion and Multi-Decoding Network for Multi-Document Reading Comprehension
Liang Wen | Houfeng Wang | Yingwei Luo | Xiaolin Wang

Multi-document reading comprehension task requires collecting evidences from different documents for answering questions. Previous research works either use the extractive modeling method to naively integrate the scores from different documents on the encoder side or use the generative modeling method to collect the clues from different documents on the decoder side individually. However, any single modeling method cannot make full of the advantages of both. In this work, we propose a novel method that tries to employ a multi-view fusion and multi-decoding mechanism to achieve it. For one thing, our approach leverages question-centered fusion mechanism and cross-attention mechanism to gather fine-grained fusion of evidence clues from different documents in the encoder and decoder concurrently. For another, our method simultaneously employs both the extractive decoding approach and the generative decoding method to effectively guide the training process. Compared with existing methods, our method can perform both extractive decoding and generative decoding independently and optionally. Our experiments on two mainstream multi-document reading comprehension datasets (Natural Questions and TriviaQA) demonstrate that our method can provide consistent improvements over previous state-of-the-art methods.

COCO-DR: Combating the Distribution Shift in Zero-Shot Dense Retrieval with Contrastive and Distributionally Robust Learning
Yue Yu | Chenyan Xiong | Si Sun | Chao Zhang | Arnold Overwijk

We present a new zero-shot dense retrieval (ZeroDR) method, COCO-DR, to improve the generalization ability of dense retrieval by combating the distribution shifts between source training tasks and target scenarios. To mitigate the impact of document differences, COCO-DR continues pretraining the language model on the target corpora to adapt the model to target distributions via COtinuous COtrastive learning. To prepare for unseen target queries, COCO-DR leverages implicit Distributionally Robust Optimization (iDRO) to reweight samples from different source query clusters for improving model robustness over rare queries during fine-tuning. COCO-DR achieves superior average performance on BEIR, the zero-shot retrieval benchmark. At BERT_Base scale, COCO-DR Base outperforms other ZeroDR models with 60x larger size. At BERT_Large scale, COCO-DR Large outperforms the giant GPT-3 embedding model which has 500x more parameters. Our analysis shows the correlation between COCO-DR’s effectiveness in combating distribution shifts and improving zero-shot accuracy. Our code and model can be found at

Language Model Pre-Training with Sparse Latent Typing
Liliang Ren | Zixuan Zhang | Han Wang | Clare Voss | ChengXiang Zhai | Heng Ji

Modern large-scale Pre-trained Language Models (PLMs) have achieved tremendous success on a wide range of downstream tasks. However, most of the LM pre-training objectives only focus on text reconstruction, but have not sought to learn latent-level interpretable representations of sentences. In this paper, we manage to push the language models to obtain a deeper understanding of sentences by proposing a new pre-training objective, Sparse Latent Typing, which enables the model to sparsely extract sentence-level keywords with diverse latent types. Experimental results show that our model is able to learn interpretable latent type categories in a self-supervised manner without using any external knowledge. Besides, the language model pre-trained with such an objective also significantly improves Information Extraction related downstream tasks in both supervised and few-shot settings. Our code is publicly available at

On the Transformation of Latent Space in Fine-Tuned NLP Models
Nadir Durrani | Hassan Sajjad | Fahim Dalvi | Firoj Alam

We study the evolution of latent space in fine-tuned NLP models. Different from the commonly used probing-framework, we opt for an unsupervised method to analyze representations. More specifically, we discover latent concepts in the representational space using hierarchical clustering. We then use an alignment function to gauge the similarity between the latent space of a pre-trained model and its fine-tuned version. We use traditional linguistic concepts to facilitate our understanding and also study how the model space transforms towards task-specific information. We perform a thorough analysis, comparing pre-trained and fine-tuned models across three models and three downstream tasks. The notable findings of our work are: i) the latent space of the higher layers evolve towards task-specific concepts, ii) whereas the lower layers retain generic concepts acquired in the pre-trained model, iii) we discovered that some concepts in the higher layers acquire polarity towards the output class, and iv) that these concepts can be used for generating adversarial triggers.

Watch the Neighbors: A Unified K-Nearest Neighbor Contrastive Learning Framework for OOD Intent Discovery
Yutao Mou | Keqing He | Pei Wang | Yanan Wu | Jingang Wang | Wei Wu | Weiran Xu

Discovering out-of-domain (OOD) intent is important for developing new skills in task-oriented dialogue systems. The key challenges lie in how to transfer prior in-domain (IND) knowledge to OOD clustering, as well as jointly learn OOD representations and cluster assignments. Previous methods suffer from in-domain overfitting problem, and there is a natural gap between representation learning and clustering objectives. In this paper, we propose a unified K-nearest neighbor contrastive learning framework to discover OOD intents. Specifically, for IND pre-training stage, we propose a KCL objective to learn inter-class discriminative features, while maintaining intra-class diversity, which alleviates the in-domain overfitting problem. For OOD clustering stage, we propose a KCC method to form compact clusters by mining true hard negative samples, which bridges the gap between clustering and representation learning. Extensive experiments on three benchmark datasets show that our method achieves substantial improvements over the state-of-the-art methods.

Extracted BERT Model Leaks More Information than You Think!
Xuanli He | Lingjuan Lyu | Chen Chen | Qiongkai Xu

The collection and availability of big data, combined with advances in pre-trained models (e.g. BERT), have revolutionized the predictive performance of natural language processing tasks. This allows corporations to provide machine learning as a service (MLaaS) by encapsulating fine-tuned BERT-based models as APIs. Due to significant commercial interest, there has been a surge of attempts to steal remote services via model extraction. Although previous works have made progress in defending against model extraction attacks, there has been little discussion on their performance in preventing privacy leakage. This work bridges this gap by launching an attribute inference attack against the extracted BERT model. Our extensive experiments reveal that model extraction can cause severe privacy leakage even when victim models are facilitated with state-of-the-art defensive strategies.

Do Vision-and-Language Transformers Learn Grounded Predicate-Noun Dependencies?
Mitja Nikolaus | Emmanuelle Salin | Stephane Ayache | Abdellah Fourtassi | Benoit Favre

Recent advances in vision-and-language modeling have seen the development of Transformer architectures that achieve remarkable performance on multimodal reasoning tasks.Yet, the exact capabilities of these black-box models are still poorly understood. While much of previous work has focused on studying their ability to learn meaning at the word-level, their ability to track syntactic dependencies between words has received less attention.We take a first step in closing this gap by creating a new multimodal task targeted at evaluating understanding of predicate-noun dependencies in a controlled setup.We evaluate a range of state-of-the-art models and find that their performance on the task varies considerably, with some models performing relatively well and others at chance level. In an effort to explain this variability, our analyses indicate that the quality (and not only sheer quantity) of pretraining data is essential. Additionally, the best performing models leverage fine-grained multimodal pretraining objectives in addition to the standard image-text matching objectives.This study highlights that targeted and controlled evaluations are a crucial step for a precise and rigorous test of the multimodal knowledge of vision-and-language models.

A Multilingual Perspective Towards the Evaluation of Attribution Methods in Natural Language Inference
Kerem Zaman | Yonatan Belinkov

Most evaluations of attribution methods focus on the English language. In this work, we present a multilingual approach for evaluating attribution methods for the Natural Language Inference (NLI) task in terms of faithfulness and plausibility.First, we introduce a novel cross-lingual strategy to measure faithfulness based on word alignments, which eliminates the drawbacks of erasure-based evaluations.We then perform a comprehensive evaluation of attribution methods, considering different output mechanisms and aggregation methods.Finally, we augment the XNLI dataset with highlight-based explanations, providing a multilingual NLI dataset with highlights, to support future exNLP studies. Our results show that attribution methods performing best for plausibility and faithfulness are different.

Graph-Based Multilingual Label Propagation for Low-Resource Part-of-Speech Tagging
Ayyoob ImaniGooghari | Silvia Severini | Masoud Jalili Sabet | François Yvon | Hinrich Schütze

Part-of-Speech (POS) tagging is an important component of the NLP pipeline, but many low-resource languages lack labeled data for training. An established method for training a POS tagger in such a scenario is to create a labeled training set by transferring from high-resource languages. In this paper, we propose a novel method for transferring labels from multiple high-resource source to low-resource target languages. We formalize POS tag projection as graph-based label propagation. Given translations of a sentence in multiple languages, we create a graph with words as nodes and alignment links as edges by aligning words for all language pairs. We then propagate node labels from source to target using a Graph Neural Network augmented with transformer layers. We show that our propagation creates training sets that allow us to train POS taggers for a diverse set of languages. When combined with enhanced contextualized embeddings, our method achieves a new state-of-the-art for unsupervised POS tagging of low-resource languages.

SubeventWriter: Iterative Sub-event Sequence Generation with Coherence Controller
Zhaowei Wang | Hongming Zhang | Tianqing Fang | Yangqiu Song | Ginny Wong | Simon See

In this paper, we propose a new task of sub-event generation for an unseen process to evaluate the understanding of the coherence of sub-event actions and objects. To solve the problem, we design SubeventWriter, a sub-event sequence generation framework with a coherence controller. Given an unseen process, the framework can iteratively construct the sub-event sequence by generating one sub-event at each iteration. We also design a very effective coherence controller to decode more coherent sub-events. As our extensive experiments and analysis indicate, SubeventWriter can generate more reliable and meaningful sub-event sequences for unseen processes.

Infinite SCAN: An Infinite Model of Diachronic Semantic Change
Seiichi Inoue | Mamoru Komachi | Toshinobu Ogiso | Hiroya Takamura | Daichi Mochihashi

In this study, we propose a Bayesian model that can jointly estimate the number of senses of words and their changes through time.The model combines a dynamic topic model on Gaussian Markov random fields with a logistic stick-breaking process that realizes Dirichlet process. In the experiments, we evaluated the proposed model in terms of interpretability, accuracy in estimating the number of senses, and tracking their changes using both artificial data and real data.We quantitatively verified that the model behaves as expected through evaluation using artificial data.Using the CCOHA corpus, we showed that our model outperforms the baseline model and investigated the semantic changes of several well-known target words.

Learning Instructions with Unlabeled Data for Zero-Shot Cross-Task Generalization
Yuxian Gu | Pei Ke | Xiaoyan Zhu | Minlie Huang

Training language models to learn from human instructions for zero-shot cross-task generalization has attracted much attention in NLP communities. Recently, instruction tuning (IT), which fine-tunes a pre-trained language model on a massive collection of tasks described via human-craft instructions, has been shown effective in instruction learning for unseen tasks. However, IT relies on a large amount of human-annotated samples, which restricts its generalization. Unlike labeled data, unlabeled data are often massive and cheap to obtain. In this work, we study how IT can be improved with unlabeled data. We first empirically explore the IT performance trends versus the number of labeled data, instructions, and training tasks. We find it critical to enlarge the number of training instructions, and the instructions can be underutilized due to the scarcity of labeled data. Then, we propose Unlabeled Data Augmented Instruction Tuning (UDIT) to take better advantage of the instructions during IT by constructing pseudo-labeled data from unlabeled plain texts. We conduct extensive experiments to show UDIT’s effectiveness in various scenarios of tasks and datasets. We also comprehensively analyze the key factors of UDIT to investigate how to better improve IT with unlabeled data. The code is publicly available at

Counterfactual Data Augmentation via Perspective Transition for Open-Domain Dialogues
Jiao Ou | Jinchao Zhang | Yang Feng | Jie Zhou

The construction of open-domain dialogue systems requires high-quality dialogue datasets. The dialogue data admits a wide variety of responses for a given dialogue history, especially responses with different semantics. However, collecting high-quality such a dataset in most scenarios is labor-intensive and time-consuming. In this paper, we propose a data augmentation method to automatically augment high-quality responses with different semantics by counterfactual inference. Specifically, given an observed dialogue, our counterfactual generation model first infers semantically different responses by replacing the observed reply perspective with substituted ones. Furthermore, our data selection method filters out detrimental augmented responses. Experimental results show that our data augmentation method can augment high-quality responses with different semantics for a given dialogue history, and can outperform competitive baselines on multiple downstream tasks.

SQUIRE: A Sequence-to-sequence Framework for Multi-hop Knowledge Graph Reasoning
Yushi Bai | Xin Lv | Juanzi Li | Lei Hou | Yincen Qu | Zelin Dai | Feiyu Xiong

Multi-hop knowledge graph (KG) reasoning has been widely studied in recent years to provide interpretable predictions on missing links with evidential paths. Most previous works use reinforcement learning (RL) based methods that learn to navigate the path towards the target entity. However, these methods suffer from slow and poor convergence, and they may fail to infer a certain path when there is a missing edge along the path. Here we present SQUIRE, the first Sequence-to-sequence based multi-hop reasoning framework, which utilizes an encoder-decoder Transformer structure to translate the query to a path. Our framework brings about two benefits: (1) It can learn and predict in an end-to-end fashion, which gives better and faster convergence; (2) Our transformer model does not rely on existing edges to generate the path, and has the flexibility to complete missing edges along the path, especially in sparse KGs. Experiments on standard and sparse KGs show that our approach yields significant improvement over prior methods, while converging 4x-7x faster.

SpeechUT: Bridging Speech and Text with Hidden-Unit for Encoder-Decoder Based Speech-Text Pre-training
Ziqiang Zhang | Long Zhou | Junyi Ao | Shujie Liu | Lirong Dai | Jinyu Li | Furu Wei

The rapid development of single-modal pre-training has prompted researchers to pay more attention to cross-modal pre-training methods. In this paper, we propose a unified-modal speech-unit-text pre-training model, SpeechUT, to connect the representations of a speech encoder and a text decoder with a shared unit encoder. Leveraging hidden-unit as an interface to align speech and text, we can decompose the speech-to-text model into a speech-to-unit model and a unit-to-text model, which can be jointly pre-trained with unpaired speech and text data respectively. Our proposed SpeechUT is fine-tuned and evaluated on automatic speech recognition (ASR) and speech translation (ST) tasks. Experimental results show that SpeechUT gets substantial improvements over strong baselines, and achieves state-of-the-art performance on both the LibriSpeech ASR and MuST-C ST tasks. To better understand the proposed SpeechUT, detailed analyses are conducted. The code and pre-trained models are available at

Learning Label Modular Prompts for Text Classification in the Wild
Hailin Chen | Amrita Saha | Shafiq Joty | Steven C.H. Hoi

Machine learning models usually assume i.i.d data during training and testing, but data and tasks in real world often change over time. To emulate the transient nature of real world, we propose a challenging but practical task: text classification in-the-wild, which introduces different non-stationary training/testing stages. Decomposing a complex task into modular components can enable robust generalisation under such non-stationary environment. However, current modular approaches in NLP do not take advantage of recent advances in parameter efficient tuning of pretrained language models. To close this gap, we propose ModularPrompt, a label-modular prompt tuning framework for text classification tasks. In ModularPrompt, the input prompt consists of a sequence of soft label prompts, each encoding modular knowledge related to the corresponding class label. In two of most formidable settings, ModularPrompt outperforms relevant baselines by a large margin demonstrating strong generalisation ability. We also conduct comprehensive analysis to validate whether the learned prompts satisfy properties of a modular representation.

Unbiased and Efficient Sampling of Dependency Trees
Miloš Stanojević

Most computational models of dependency syntax consist of distributions over spanning trees. However, the majority of dependency treebanks require that every valid dependency tree has a single edge coming out of the ROOT node, a constraint that is not part of the definition of spanning trees. For this reason all standard inference algorithms for spanning trees are sub-optimal for inference over dependency trees.Zmigrod et al (2021) proposed algorithms for sampling with and without replacement from the dependency tree distribution that incorporate the single-root constraint. In this paper we show that their fastest algorithm for sampling with replacement, Wilson-RC, is in fact producing biased samples and we provide two alternatives that are unbiased. Additionally, we propose two algorithms (one incremental, one parallel) that reduce the asymptotic runtime of algorithm for sampling k trees without replacement to O(kn^3). These algorithms are both asymptotically and practically more efficient.

Continual Learning of Neural Machine Translation within Low Forgetting Risk Regions
Shuhao Gu | Bojie Hu | Yang Feng

This paper considers continual learning of large-scale pretrained neural machine translation model without accessing the previous training data or introducing model separation. We argue that the widely used regularization-based methods, which perform multi-objective learning with an auxiliary loss, suffer from the misestimate problem and cannot always achieve a good balance between the previous and new tasks. To solve the problem, we propose a two-stage training method based on the local features of the real loss. We first search low forgetting risk regions, where the model can retain the performance on the previous task as the parameters are updated, to avoid the catastrophic forgetting problem. Then we can continually train the model within this region only with the new training data to fit the new task. Specifically, we propose two methods to search the low forgetting risk regions, which are based on the curvature of loss and the impacts of the parameters on the model output, respectively. We conduct experiments on domain adaptation and more challenging language adaptation tasks, and the experimental results show that our method can achieve significant improvements compared with several strong baselines.

COST-EFF: Collaborative Optimization of Spatial and Temporal Efficiency with Slenderized Multi-exit Language Models
Bowen Shen | Zheng Lin | Yuanxin Liu | Zhengxiao Liu | Lei Wang | Weiping Wang

Transformer-based pre-trained language models (PLMs) mostly suffer from excessive overhead despite their advanced capacity. For resource-constrained devices, there is an urgent need for a spatially and temporally efficient model which retains the major capacity of PLMs. However, existing statically compressed models are unaware of the diverse complexities between input instances, potentially resulting in redundancy and inadequacy for simple and complex inputs. Also, miniature models with early exiting encounter challenges in the trade-off between making predictions and serving the deeper layers. Motivated by such considerations, we propose a collaborative optimization for PLMs that integrates static model compression and dynamic inference acceleration. Specifically, the PLM is slenderized in width while the depth remains intact, complementing layer-wise early exiting to speed up inference dynamically. To address the trade-off of early exiting, we propose a joint training approach that calibrates slenderization and preserves contributive structures to each exit instead of only the final layer. Experiments are conducted on GLUE benchmark and the results verify the Pareto optimality of our approach at high compression and acceleration rate with 1/8 parameters and 1/19 FLOPs of BERT.

Rescue Implicit and Long-tail Cases: Nearest Neighbor Relation Extraction
Zhen Wan | Qianying Liu | Zhuoyuan Mao | Fei Cheng | Sadao Kurohashi | Jiwei Li

Relation extraction (RE) has achieved remarkable progress with the help of pre-trained language models. However, existing RE models are usually incapable of handling two situations: implicit expressions and long-tail relation types, caused by language complexity and data sparsity. In this paper, we introduce a simple enhancement of RE using k nearest neighbors (kNN-RE). kNN-RE allows the model to consult training relations at test time through a nearest-neighbor search and provides a simple yet effective means to tackle the two issues above. Additionally, we observe that kNN-RE serves as an effective way to leverage distant supervision (DS) data for RE. Experimental results show that the proposed kNN-RE achieves state-of-the-art performances on a variety of supervised RE datasets, i.e., ACE05, SciERC, and Wiki80, along with outperforming the best model to date on the i2b2 and Wiki80 datasets in the setting of allowing using DS. Our code and models are available at:

StoryER: Automatic Story Evaluation via Ranking, Rating and Reasoning
Hong Chen | Duc Vo | Hiroya Takamura | Yusuke Miyao | Hideki Nakayama

Existing automatic story evaluation methods place a premium on story lexical level coherence, deviating from human preference.We go beyond this limitation by considering a novel Story Evaluation method that mimics human preference when judging a story, namely StoryER, which consists of three sub-tasks: Ranking, Rating and Reasoning.Given either a machine-generated or a human-written story, StoryER requires the machine to output 1) a preference score that corresponds to human preference, 2) specific ratings and their corresponding confidences and 3) comments for various aspects (e.g., opening, character-shaping).To support these tasks, we introduce a well-annotated dataset comprising (i) 100k ranked story pairs; and (ii) a set of 46k ratings and comments on various aspects of the story.We finetune Longformer-Encoder-Decoder (LED) on the collected dataset, with the encoder responsible for preference score and aspect prediction and the decoder for comment generation.Our comprehensive experiments result a competitive benchmark for each task, showing the high correlation to human preference.In addition, we have witnessed the joint learning of the preference scores, the aspect ratings, and the comments brings gain each single task.Our dataset and benchmarks are publicly available to advance the research of story evaluation tasks.

Enhancing Self-Consistency and Performance of Pre-Trained Language Models through Natural Language Inference
Eric Mitchell | Joseph Noh | Siyan Li | Will Armstrong | Ananth Agarwal | Patrick Liu | Chelsea Finn | Christopher Manning

While large pre-trained language models are powerful, their predictions often lack logical consistency across test inputs. For example, a state-of-the-art Macaw question-answering (QA) model answers <i>Yes</i> to <i>Is a sparrow a bird?</i> and <i>Does a bird have feet?</i> but answers <i>No</i> to <i>Does a sparrow have feet?</i>. To address this failure mode, we propose a framework, Consistency Correction through Relation Detection, or <b>ConCoRD</b>, for boosting the consistency and accuracy of pre-trained NLP models using pre-trained natural language inference (NLI) models without fine-tuning or re-training. Given a batch of test inputs, ConCoRD samples several candidate outputs for each input and instantiates a factor graph that accounts for both the model’s belief about the likelihood of each answer choice in isolation and the NLI model’s beliefs about pair-wise answer choice compatibility. We show that a weighted MaxSAT solver can efficiently compute high-quality answer choices under this factor graph, improving over the raw model’s predictions. Our experiments demonstrate that ConCoRD consistently boosts accuracy and consistency of off-the-shelf closed-book QA and VQA models using off-the-shelf NLI models, notably increasing accuracy of LXMERT on ConVQA by 5% absolute. See the project website ( for code and data.

Robustness of Demonstration-based Learning Under Limited Data Scenario
Hongxin Zhang | Yanzhe Zhang | Ruiyi Zhang | Diyi Yang

Demonstration-based learning has shown great potential in stimulating pretrained language models’ ability under limited data scenario. Simply augmenting the input with some demonstrations can significantly improve performance on few-shot NER. However, why such demonstrations are beneficial for the learning process remains unclear since there is no explicit alignment between the demonstrations and the predictions. In this paper, we design pathological demonstrations by gradually removing intuitively useful information from the standard ones to take a deep dive of the robustness of demonstration-based sequence labeling and show that (1) demonstrations composed of random tokens still make the model a better few-shot learner; (2) the length of random demonstrations and the relevance of random tokens are the main factors affecting the performance; (3) demonstrations increase the confidence of model predictions on captured superficial patterns. We have publicly released our code at

Modeling Information Change in Science Communication with Semantically Matched Paraphrases
Dustin Wright | Jiaxin Pei | David Jurgens | Isabelle Augenstein

Whether the media faithfully communicate scientific information has long been a core issue to the science community. Automatically identifying paraphrased scientific findings could enable large-scale tracking and analysis of information changes in the science communication process, but this requires systems to understand the similarity between scientific information across multiple domains. To this end, we present the SCIENTIFIC PARAPHRASE AND INFORMATION CHANGE DATASET (SPICED), the first paraphrase dataset of scientific findings annotated for degree of information change. SPICED contains 6,000 scientific finding pairs extracted from news stories, social media discussions, and full texts of original papers. We demonstrate that SPICED poses a challenging task and that models trained on SPICED improve downstream performance on evidence retrieval for fact checking of real-world scientific claims. Finally, we show that models trained on SPICED can reveal large-scale trends in the degrees to which people and organizations faithfully communicate new scientific findings. Data, code, and pre-trained models are available at

Word Order Matters When You Increase Masking
Karim Lasri | Alessandro Lenci | Thierry Poibeau

Word order, an essential property of natural languages, is injected in Transformer-based neural language models using position encoding. However, recent experiments have shown that explicit position encoding is not always useful, since some models without such feature managed to achieve state-of-the art performance on some tasks. To understand better this phenomenon, we examine the effect of removing position encodings on the pre-training objective itself (i.e., masked language modelling), to test whether models can reconstruct position information from co-occurrences alone. We do so by controlling the amount of masked tokens in the input sentence, as a proxy to affect the importance of position information for the task. We find that the necessity of position information increases with the amount of masking, and that masked language models without position encodings are not able to reconstruct this information on the task. These findings point towards a direct relationship between the amount of masking and the ability of Transformers to capture order-sensitive aspects of language using position encoding.

An Empirical Analysis of Memorization in Fine-tuned Autoregressive Language Models
Fatemehsadat Mireshghallah | Archit Uniyal | Tianhao Wang | David Evans | Taylor Berg-Kirkpatrick

Large language models are shown to present privacy risks through memorization of training data, andseveral recent works have studied such risks for the pre-training phase. Little attention, however, has been given to the fine-tuning phase and it is not well understood how different fine-tuning methods (such as fine-tuning the full model, the model head, and adapter) compare in terms of memorization risk. This presents increasing concern as the “pre-train and fine-tune” paradigm proliferates. In this paper, we empirically study memorization of fine-tuning methods using membership inference and extraction attacks, and show that their susceptibility to attacks is very different. We observe that fine-tuning the head of the model has the highest susceptibility to attacks, whereas fine-tuning smaller adapters appears to be less vulnerable to known extraction attacks.

Style Transfer as Data Augmentation: A Case Study on Named Entity Recognition
Shuguang Chen | Leonardo Neves | Thamar Solorio

In this work, we take the named entity recognition task in the English language as a case study and explore style transfer as a data augmentation method to increase the size and diversity of training data in low-resource scenarios. We propose a new method to effectively transform the text from a high-resource domain to a low-resource domain by changing its style-related attributes to generate synthetic data for training. Moreover, we design a constrained decoding algorithm along with a set of key ingredients for data selection to guarantee the generation of valid and coherent data. Experiments and analysis on five different domain pairs under different data regimes demonstrate that our approach can significantly improve results compared to current state-of-the-art data augmentation methods. Our approach is a practical solution to data scarcity, and we expect it to be applicable to other NLP tasks.

Linguistic Corpus Annotation for Automatic Text Simplification Evaluation
Rémi Cardon | Adrien Bibal | Rodrigo Wilkens | David Alfter | Magali Norré | Adeline Müller | Watrin Patrick | Thomas François

Evaluating automatic text simplification (ATS) systems is a difficult task that is either performed by automatic metrics or user-based evaluations. However, from a linguistic point-of-view, it is not always clear on what bases these evaluations operate. In this paper, we propose annotations of the ASSET corpus that can be used to shed more light on ATS evaluation. In addition to contributing with this resource, we show how it can be used to analyze SARI’s behavior and to re-evaluate existing ATS systems. We present our insights as a step to improve ATS evaluation protocols in the future.

Semantic Framework based Query Generation for Temporal Question Answering over Knowledge Graphs
Wentao Ding | Hao Chen | Huayu Li | Yuzhong Qu

Answering factual questions with temporal intent over knowledge graphs (temporal KGQA) attracts rising attention in recent years.In the generation of temporal queries, existing KGQA methods ignore the fact that some intrinsic connections between events can make them temporally related, which may limit their capability.We systematically analyze the possible interpretation of temporal constraints and conclude the interpretation structures as the Semantic Framework of Temporal Constraints, SF-TCons.Based on the semantic framework, we propose a temporal question answering method, SF-TQA, which generates query graphs by exploring the relevant facts of mentioned entities, where the exploring process is restricted by SF-TCons. Our evaluations show that SF-TQA significantly outperforms existing methods on two benchmarks over different knowledge graphs.

There Is No Standard Answer: Knowledge-Grounded Dialogue Generation with Adversarial Activated Multi-Reference Learning
Xueliang Zhao | Tingchen Fu | Chongyang Tao | Rui Yan

Knowledge-grounded dialogue (KGC) shows excellent potential to deliver an engaging and informative response. However, existing approaches emphasize selecting one golden knowledge given a particular dialogue context, overlooking the one-to-many phenomenon in dialogue. As a result, existing paradigm limits the diversity of knowledge selection and generation. To this end, we establish a multi-reference KGC dataset and propose a series of metrics to systematically assess the one-to-many efficacy of existing KGC models. Furthermore, to extend the hypothesis space of knowledge selection to enhance the mapping relationship between multiple knowledge and multiple responses, we devise a span-based variational model and optimize the model in a wake-sleep style with an ameliorated evidence lower bound objective to learn the one-to-many generalization. Both automatic and human evaluations demonstrate the efficacy of our approach.

Stop Measuring Calibration When Humans Disagree
Joris Baan | Wilker Aziz | Barbara Plank | Raquel Fernandez

Calibration is a popular framework to evaluate whether a classifier knows when it does not know - i.e., its predictive probabilities are a good indication of how likely a prediction is to be correct. Correctness is commonly estimated against the human majority class. Recently, calibration to human majority has been measured on tasks where humans inherently disagree about which class applies. We show that measuring calibration to human majority given inherent disagreements is theoretically problematic, demonstrate this empirically on the ChaosNLI dataset, and derive several instance-level measures of calibration that capture key statistical properties of human judgements - including class frequency, ranking and entropy.

Improving compositional generalization for multi-step quantitative reasoning in question answering
Armineh Nourbakhsh | Cathy Jiao | Sameena Shah | Carolyn Rosé

Quantitative reasoning is an important aspect of question answering, especially when numeric and verbal cues interact to indicate sophisticated, multi-step programs. In this paper, we demonstrate how modeling the compositional nature of quantitative text can enhance the performance and robustness of QA models, allowing them to capture arithmetic logic that is expressed verbally. Borrowing from the literature on semantic parsing, we propose a method that encourages the QA models to adjust their attention patterns and capture input/output alignments that are meaningful to the reasoning task. We show how this strategy improves program accuracy and renders the models more robust against overfitting as the number of reasoning steps grows. Our approach is designed as a standalone module which can be prepended to many existing models and trained in an end-to-end fashion without the need for additional supervisory signal. As part of this exercise, we also create a unified dataset building on four previously released numerical QA datasets over tabular data.

A Comprehensive Comparison of Neural Networks as Cognitive Models of Inflection
Adam Wiemerslage | Shiran Dudy | Katharina Kann

Neural networks have long been at the center of a debate around the cognitive mechanism by which humans process inflectional morphology. This debate has gravitated into NLP by way of the question: Are neural networks a feasible account for human behavior in morphological inflection?We address that question by measuring the correlation between human judgments and neural network probabilities for unknown word inflections. We test a larger range of architectures than previously studied on two important tasks for the cognitive processing debate: English past tense, and German number inflection. We find evidence that the Transformer may be a better account of human behavior than LSTMs on these datasets, and that LSTM features known to increase inflection accuracy do not always result in more human-like behavior.

Can Visual Context Improve Automatic Speech Recognition for an Embodied Agent?
Pradip Pramanick | Chayan Sarkar

The usage of automatic speech recognition (ASR) systems are becoming omnipresent ranging from personal assistant to chatbots, home, and industrial automation systems, etc. Modern robots are also equipped with ASR capabilities for interacting with humans as speech is the most natural interaction modality. However, ASR in robots faces additional challenges as compared to a personal assistant. Being an embodied agent, a robot must recognize the physical entities around it and therefore reliably recognize the speech containing the description of such entities. However, current ASR systems are often unable to do so due to limitations in ASR training, such as generic datasets and open-vocabulary modeling. Also, adverse conditions during inference, such as noise, accented, and far-field speech makes the transcription inaccurate. In this work, we present a method to incorporate a robot’s visual information into an ASR system and improve the recognition of a spoken utterance containing a visible entity. Specifically, we propose a new decoder biasing technique to incorporate the visual context while ensuring the ASR output does not degrade for incorrect context. We achieve a 59% relative reduction in WER from an unmodified ASR system.

AfroLID: A Neural Language Identification Tool for African Languages
Ife Adebara | AbdelRahim Elmadany | Muhammad Abdul-Mageed | Alcides Inciarte

Language identification (LID) is a crucial precursor for NLP, especially for mining web data. Problematically, most of the world’s 7000+ languages today are not covered by LID technologies. We address this pressing issue for Africa by introducing AfroLID, a neural LID toolkit for 517 African languages and varieties. AfroLID exploits a multi-domain web dataset manually curated from across 14 language families utilizing five orthographic systems. When evaluated on our blind Test set, AfroLID achieves 95.89 F_1-score. We also compare AfroLID to five existing LID tools that each cover a small number of African languages, finding it to outperform them on most languages. We further show the utility of AfroLID in the wild by testing it on the acutely under-served Twitter domain. Finally, we offer a number of controlled case studies and perform a linguistically-motivated error analysis that allow us to both showcase AfroLID’s powerful capabilities and limitations

EvEntS ReaLM: Event Reasoning of Entity States via Language Models
Evangelia Spiliopoulou | Artidoro Pagnoni | Yonatan Bisk | Eduard Hovy

This paper investigates models of event implications. Specifically, how well models predict entity state-changes, by targeting their understanding of physical attributes. Nominally, Large Language models (LLM) have been exposed to procedural knowledge about how objects interact, yet our benchmarking shows they fail to reason about the world. Conversely, we also demonstrate that existing approaches often misrepresent the surprising abilities of LLMs via improper task encodings and that proper model prompting can dramatically improve performance of reported baseline results across multiple tasks. In particular, our results indicate that our prompting technique is especially useful for unseen attributes (out-of-domain) or when only limited data is available.

Large language models are few-shot clinical information extractors
Monica Agrawal | Stefan Hegselmann | Hunter Lang | Yoon Kim | David Sontag

A long-running goal of the clinical NLP community is the extraction of important variables trapped in clinical notes. However, roadblocks have included dataset shift from the general domain and a lack of public clinical corpora and annotations. In this work, we show that large language models, such as InstructGPT (Ouyang et al., 2022), perform well at zero- and few-shot information extraction from clinical text despite not being trained specifically for the clinical domain. Whereas text classification and generation performance have already been studied extensively in such models, here we additionally demonstrate how to leverage them to tackle a diverse set of NLP tasks which require more structured outputs, including span identification, token-level sequence classification, and relation extraction. Further, due to the dearth of available data to evaluate these systems, we introduce new datasets for benchmarking few-shot clinical information extraction based on a manual re-annotation of the CASI dataset (Moon et al., 2014) for new tasks. On the clinical extraction tasks we studied, the GPT-3 systems significantly outperform existing zero- and few-shot baselines.

Towards a Unified Multi-Dimensional Evaluator for Text Generation
Ming Zhong | Yang Liu | Da Yin | Yuning Mao | Yizhu Jiao | Pengfei Liu | Chenguang Zhu | Heng Ji | Jiawei Han

Multi-dimensional evaluation is the dominant paradigm for human evaluation in Natural Language Generation (NLG), i.e., evaluating the generated text from multiple explainable dimensions, such as coherence and fluency. However, automatic evaluation in NLG is still dominated by similarity-based metrics, and we lack a reliable framework for a more comprehensive evaluation of advanced models. In this paper, we propose a unified multi-dimensional evaluator UniEval for NLG. We re-frame NLG evaluation as a Boolean Question Answering (QA) task, and by guiding the model with different questions, we can use one evaluator to evaluate from multiple dimensions. Furthermore, thanks to the unified Boolean QA format, we are able to introduce an intermediate learning phase that enables UniEval to incorporate external knowledge from multiple related tasks and gain further improvement. Experiments on three typical NLG tasks show that UniEval correlates substantially better with human judgments than existing metrics. Specifically, compared to the top-performing unified evaluators, UniEval achieves a 23% higher correlation on text summarization, and over 43% on dialogue response generation. Also, UniEval demonstrates a strong zero-shot learning ability for unseen evaluation dimensions and tasks. Source code, data, and all pre-trained evaluators are available at

GeoMLAMA: Geo-Diverse Commonsense Probing on Multilingual Pre-Trained Language Models
Da Yin | Hritik Bansal | Masoud Monajatipoor | Liunian Harold Li | Kai-Wei Chang

Recent work has shown that Pre-trained Language Models (PLMs) store the relational knowledge learned from data and utilize it for performing downstream tasks. However, commonsense knowledge across different regions may vary. For instance, the color of bridal dress is white in American weddings whereas it is red in Chinese weddings. In this paper, we introduce a benchmark dataset, Geo-diverse Commonsense Multilingual Language Models Analysis (GeoMLAMA), for probing the diversity of the relational knowledge in multilingual PLMs. GeoMLAMA contains 3125 prompts in English, Chinese, Hindi, Persian, and Swahili, with a wide coverage of concepts shared by people from American, Chinese, Indian, Iranian and Kenyan cultures. We benchmark 11 standard multilingual PLMs on GeoMLAMA. Interestingly, we find that 1) larger multilingual PLMs variants do not necessarily store geo-diverse concepts better than its smaller variant; 2) multilingual PLMs are not intrinsically biased towards knowledge from the Western countries (the United States); 3) the native language of a country may not be the best language to probe its knowledge and 4) a language may better probe knowledge about a non-native country than its native country.

The (Undesired) Attenuation of Human Biases by Multilinguality
Cristina España-Bonet | Alberto Barrón-Cedeño

Some human preferences are universal. The odor of vanilla is perceived as pleasant all around the world. We expect neural models trained on human texts to exhibit these kind of preferences, i.e. biases, but we show that this is not always the case. We explore 16 static and contextual embedding models in 9 languages and, when possible, compare them under similar training conditions. We introduce and release CA-WEAT, multilingual cultural aware tests to quantify biases, and compare them to previous English-centric tests. Our experiments confirm that monolingual static embeddings do exhibit human biases, but values differ across languages, being far from universal. Biases are less evident in contextual models, to the point that the original human association might be reversed. Multilinguality proves to be another variable that attenuates and even reverses the effect of the bias, specially in contextual multilingual models. In order to explain this variance among models and languages, we examine the effect of asymmetries in the training corpus, departures from isomorphism in multilingual embedding spaces and discrepancies in the testing measures between languages.

Entailer: Answering Questions with Faithful and Truthful Chains of Reasoning
Oyvind Tafjord | Bhavana Dalvi Mishra | Peter Clark

Our goal is a question-answering (QA) system that can show how its answers are implied by its own internal beliefs via a systematic chain of reasoning. Such a capability would allow better understanding of why a model produced the answer it did. Our approach is to recursively combine a trained backward-chainingmodel, capable of generating a set of premises entailing an answer hypothesis, with a verifier that checks that the model itself believes those premises (and the entailment itself) through self-querying. To our knowledge, this is the first system to generate multistep chains that are both faithful (the answer follows from the reasoning) and truthful (the chain reflects the system’s own internal beliefs). In evaluation using two different datasets, users judge that a majority (70%+) of generated chains clearly show how an answer follows from a set of facts - substantially better than a high-performance baseline - while preserving answer accuracy. By materializing model beliefs that systematically support an answer, new opportunities arise for understanding the model’s system of belief, and diagnosing and correcting its misunderstandings when an answer is wrong.

Near-Negative Distinction: Giving a Second Life to Human Evaluation Datasets
Philippe Laban | Chien-Sheng Wu | Wenhao Liu | Caiming Xiong

Precisely assessing the progress in natural language generation (NLG) tasks is challenging, and human evaluation to establish a preference in a model’s output over another is often necessary.However, human evaluation is usually costly, difficult to reproduce, and non-reusable.In this paper, we propose a new and simple automatic evaluation method for NLG called Near-Negative Distinction (NND) that repurposes prior human annotations into NND tests.In an NND test, an NLG model must place a higher likelihood on a high-quality output candidate than on a near-negative candidate with a known error.Model performance is established by the number of NND tests a model passes, as well as the distribution over task-specific errors the model fails on.Through experiments on three NLG tasks (question generation, question answering, and summarization), we show that NND achieves a higher correlation with human judgments than standard NLG evaluation metrics. We then illustrate NND evaluation in four practical scenarios, for example performing fine-grain model analysis, or studying model training dynamics. Our findings suggest that NND can give a second life to human annotations and provide low-cost NLG evaluation.

ToKen: Task Decomposition and Knowledge Infusion for Few-Shot Hate Speech Detection
Badr AlKhamissi | Faisal Ladhak | Srinivasan Iyer | Veselin Stoyanov | Zornitsa Kozareva | Xian Li | Pascale Fung | Lambert Mathias | Asli Celikyilmaz | Mona Diab

Hate speech detection is complex; it relies on commonsense reasoning, knowledge of stereotypes, and an understanding of social nuance that differs from one culture to the next. It is also difficult to collect a large-scale hate speech annotated dataset. In this work, we frame this problem as a few-shot learning task, and show significant gains with decomposing the task into its “constituent” parts. In addition, we see that infusing knowledge from reasoning datasets (e.g. ATOMIC2020) improves the performance even further. Moreover, we observe that the trained models generalize to out-of-distribution datasets, showing the superiority of task decomposition and knowledge infusion compared to previously used methods. Concretely, our method outperforms the baseline by 17.83% absolute gain in the 16-shot case.

Are Hard Examples also Harder to Explain? A Study with Human and Model-Generated Explanations
Swarnadeep Saha | Peter Hase | Nazneen Rajani | Mohit Bansal

Recent work on explainable NLP has shown that few-shot prompting can enable large pre-trained language models (LLMs) to generate grammatical and factual natural language explanations for data labels. In this work, we study the connection between explainability and sample hardness by investigating the following research question – “Are LLMs and humans equally good at explaining data labels for both easy and hard samples?” We answer this question by first collecting human-written explanations in the form of generalizable commonsense rules on the task of Winograd Schema Challenge (Winogrande dataset). We compare these explanations with those generated by GPT-3 while varying the hardness of the test samples as well as the in-context samples. We observe that (1) GPT-3 explanations are as grammatical as human explanations regardless of the hardness of the test samples, (2) for easy examples, GPT-3 generates highly supportive explanations but human explanations are more generalizable, and (3) for hard examples, human explanations are significantly better than GPT-3 explanations both in terms of label-supportiveness and generalizability judgements. We also find that hardness of the in-context examples impacts the quality of GPT-3 explanations. Finally, we show that the supportiveness and generalizability aspects of human explanations are also impacted by sample hardness, although by a much smaller margin than models.

Stanceosaurus: Classifying Stance Towards Multicultural Misinformation
Jonathan Zheng | Ashutosh Baheti | Tarek Naous | Wei Xu | Alan Ritter

We present Stanceosaurus, a new corpus of 28,033 tweets in English, Hindi and Arabic annotated with stance towards 250 misinformation claims. As far as we are aware, it is the largest corpus annotated with stance towards misinformation claims. The claims in Stanceosaurus originate from 15 fact-checking sources that cover diverse geographical regions and cultures. Unlike existing stance datasets, we introduce a more fine-grained 5-class labeling strategy with additional subcategories to distinguish implicit stance. Pre-trained transformer-based stance classifiers that are fine-tuned on our corpus show good generalization on unseen claims and regional claims from countries outside the training data. Cross-lingual experiments demonstrate Stanceosaurus’ capability of training multilingual models, achieving 53.1 F1 on Hindi and 50.4 F1 on Arabic without any target-language fine-tuning. Finally, we show how a domain adaptation method can be used to improve performance on Stanceosaurus using additional RumourEval-2019 data. We will make Stanceosaurus publicly available to the research community upon publication and hope it will encourage further work on misinformation identification across languages and cultures.

Gendered Mental Health Stigma in Masked Language Models
Inna Lin | Lucille Njoo | Anjalie Field | Ashish Sharma | Katharina Reinecke | Tim Althoff | Yulia Tsvetkov

Mental health stigma prevents many individuals from receiving the appropriate care, and social psychology studies have shown that mental health tends to be overlooked in men. In this work, we investigate gendered mental health stigma in masked language models. In doing so, we operationalize mental health stigma by developing a framework grounded in psychology research: we use clinical psychology literature to curate prompts, then evaluate the models’ propensity to generate gendered words. We find that masked language models capture societal stigma about gender in mental health: models are consistently more likely to predict female subjects than male in sentences about having a mental health condition (32% vs. 19%), and this disparity is exacerbated for sentences that indicate treatment-seeking behavior. Furthermore, we find that different models capture dimensions of stigma differently for men and women, associating stereotypes like anger, blame, and pity more with women with mental health conditions than with men. In showing the complex nuances of models’ gendered mental health stigma, we demonstrate that context and overlapping dimensions of identity are important considerations when assessing computational models’ social biases.

Efficient Nearest Neighbor Search for Cross-Encoder Models using Matrix Factorization
Nishant Yadav | Nicholas Monath | Rico Angell | Manzil Zaheer | Andrew McCallum

Efficient k-nearest neighbor search is a fundamental task, foundational for many problems in NLP. When the similarity is measured by dot-product between dual-encoder vectors or L2-distance, there already exist many scalable and efficient search methods. But not so when similarity is measured by more accurate and expensive black-box neural similarity models, such as cross-encoders, which jointly encode the query and candidate neighbor. The cross-encoders’ high computational cost typically limits their use to reranking candidates retrieved by a cheaper model, such as dual encoder or TF-IDF. However, the accuracy of such a two-stage approach is upper-bounded by the recall of the initial candidate set, and potentially requires additional training to align the auxiliary retrieval model with the cross-encoder model. In this paper, we present an approach that avoids the use of a dual-encoder for retrieval, relying solely on the cross-encoder. Retrieval is made efficient with CUR decomposition, a matrix decomposition approach that approximates all pairwise cross-encoder distances from a small subset of rows and columns of the distance matrix. Indexing items using our approach is computationally cheaper than training an auxiliary dual-encoder model through distillation. Empirically, for k > 10, our approach provides test-time recall-vs-computational cost trade-offs superior to the current widely-used methods that re-rank items retrieved using a dual-encoder or TF-IDF.

Prompt-and-Rerank: A Method for Zero-Shot and Few-Shot Arbitrary Textual Style Transfer with Small Language Models
Mirac Suzgun | Luke Melas-Kyriazi | Dan Jurafsky

We propose a method for arbitrary textual style transfer (TST)—the task of transforming a text into any given style—utilizing general-purpose pre-trained language models. Our method, Prompt-and-Rerank, is based on a mathematical formulation of the TST task, decomposing it into three constituent components: textual similarity, target style strength, and fluency. Our method uses zero-shot or few-shot prompting to obtain a set of candidate generations in the target style, and then re-ranks them according to the three components. Our method enables small pre-trained language models to perform on par with state-of-the-art large-scale models while using two orders of magnitude less compute and memory. We also investigate the effect of model size and prompt design (e.g., prompt paraphrasing and delimiter-pair choice) on style transfer quality across seven diverse textual style transfer datasets, finding, among other things, that delimiter-pair choice has a large impact on performance, and that models have biases on the direction of style transfer.

Learning to Decompose: Hypothetical Question Decomposition Based on Comparable Texts
Ben Zhou | Kyle Richardson | Xiaodong Yu | Dan Roth

Explicit decomposition modeling, which involves breaking down complex tasks into more straightforward and often more interpretable sub-tasks, has long been a central theme in developing robust and interpretable NLU systems. However, despite the many datasets and resources built as part of this effort, the majority have small-scale annotations and limited scope, which is insufficient to solve general decomposition tasks. In this paper, we look at large-scale intermediate pre-training of decomposition-based transformers using distant supervision from comparable texts, particularly large-scale parallel news. We show that with such intermediate pre-training, developing robust decomposition-based models for a diverse range of tasks becomes more feasible. For example, on semantic parsing, our model, DecompT5, improves 20% to 30% on two datasets, Overnight and TORQUE, over the baseline language model. We further use DecompT5 to build a novel decomposition-based QA system named DecompEntail, improving over state-of-the-art models, including GPT-3, on both HotpotQA and StrategyQA by 8% and 4%, respectively.

Why is Winoground Hard? Investigating Failures in Visuolinguistic Compositionality
Anuj Diwan | Layne Berry | Eunsol Choi | David Harwath | Kyle Mahowald

Recent visuolinguistic pre-trained models show promising progress on various end tasks such as image retrieval and video captioning. Yet, they fail miserably on the recently proposed Winoground dataset, which challenges models to match paired images and English captions, with items constructed to overlap lexically but differ in meaning (e.g., “there is a mug in some grass” vs. “there is some grass in a mug”). By annotating the dataset using new fine-grained tags, we show that solving the Winoground task requires not just compositional language understanding, but a host of other abilities like commonsense reasoning or locating small, out-of-focus objects in low-resolution images. In this paper, we identify the dataset’s main challenges through a suite of experiments on related tasks (probing task, image retrieval task), data augmentation, and manual inspection of the dataset. Our analysis suggests that a main challenge in visuolinguistic models may lie in fusing visual and textual representations, rather than in compositional language understanding. We release our annotation and code at

Gradient-based Constrained Sampling from Language Models
Sachin Kumar | Biswajit Paria | Yulia Tsvetkov

Large pretrained language models are successful at generating fluent text but are notoriously hard to controllably sample from. In this work, we study constrained sampling from such language models, i.e., generating text that satisfies user-defined constraints, while maintaining fluency and model’s performance in a downstream task. We propose MuCoLa—a sampling procedure that combines the log-likelihood of the language model with arbitrary (differentiable) constraints in a single energy function, and then generates samples in a non-autoregressive manner. Specifically, it initializes the entire output sequence with noise and follows a Markov chain defined by Langevin Dynamics using the gradients of this energy. We evaluate MuCoLa on text generation with soft and hard constraints as well as their combinations, obtaining significant improvements over competitive baselines for toxicity avoidance, sentiment control, and keyword-guided generation.

TaCube: Pre-computing Data Cubes for Answering Numerical-Reasoning Questions over Tabular Data
Fan Zhou | Mengkang Hu | Haoyu Dong | Zhoujun Cheng | Fan Cheng | Shi Han | Dongmei Zhang

Existing auto-regressive pre-trained language models (PLMs) like T5 and BART, have been well applied to table question answering by UNIFIEDSKG and TAPEX, respectively, and demonstrated state-of-the-art results on multiple benchmarks. However, auto-regressive PLMs are challenged by recent emerging numerical reasoning datasets, such as TAT-QA, due to the error-prone implicit calculation. In this paper, we present TaCube, to pre-compute aggregation/arithmetic results for the table in advance, so that they are handy and readily available for PLMs to answer numerical reasoning questions. TaCube systematically and comprehensively covers a collection of computational operations over table segments. By simply concatenating TaCube to the input sequence of PLMs, it shows significant experimental effectiveness. TaCube promotes the F1 score from 49.6% to 66.2% on TAT-QA and achieves new state-of-the-art results on WikiTQ (59.6% denotation accuracy). TaCube’s improvements on numerical reasoning cases are even more notable: on TAT-QA, TaCube promotes the exact match accuracy of BART-large by 39.6% on sum, 52.5% on average, 36.6% on substraction, and 22.2% on division. We believe that TaCube is a general and portable pre-computation solution that can be potentially integrated to various numerical reasoning frameworks

Rich Knowledge Sources Bring Complex Knowledge Conflicts: Recalibrating Models to Reflect Conflicting Evidence
Hung-Ting Chen | Michael Zhang | Eunsol Choi

Question answering models can use rich knowledge sources — up to one hundred retrieved passages and parametric knowledge in the large-scale language model (LM). Prior work assumes information in such knowledge sources is consistent with each other, paying little attention to how models blend information stored in their LM parameters with that from retrieved evidence documents. In this paper, we simulate knowledge conflicts (i.e., where parametric knowledge suggests one answer and different passages suggest different answers) and examine model behaviors. We find retrieval performance heavily impacts which sources models rely on, and current models mostly rely on non-parametric knowledgein their best-performing settings. We discover a troubling trend that contradictions among knowledge sources affect model confidence only marginally. To address this issue, we present a new calibration study, where models are discouraged from presenting any single answer when presented with multiple conflicting answer candidates in retrieved evidences.

QA Domain Adaptation using Hidden Space Augmentation and Self-Supervised Contrastive Adaptation
Zhenrui Yue | Huimin Zeng | Bernhard Kratzwald | Stefan Feuerriegel | Dong Wang

Question answering (QA) has recently shown impressive results for answering questions from customized domains. Yet, a common challenge is to adapt QA models to an unseen target domain. In this paper, we propose a novel self-supervised framework called QADA for QA domain adaptation. QADA introduces a novel data augmentation pipeline used to augment training QA samples. Different from existing methods, we enrich the samples via hidden space augmentation. For questions, we introduce multi-hop synonyms and sample augmented token embeddings with Dirichlet distributions. For contexts, we develop an augmentation method which learns to drop context spans via a custom attentive sampling strategy. Additionally, contrastive learning is integrated in the proposed self-supervised adaptation framework QADA. Unlike existing approaches, we generate pseudo labels and propose to train the model via a novel attention-based contrastive adaptation method. The attention weights are used to build informative features for discrepancy estimation that helps the QA model separate answers and generalize across source and target domains. To the best of our knowledge, our work is the first to leverage hidden space augmentation and attention-based contrastive adaptation for self-supervised domain adaptation in QA. Our evaluation shows that QADA achieves considerable improvements on multiple target datasets over state-of-the-art baselines in QA domain adaptation.

When FLUE Meets FLANG: Benchmarks and Large Pretrained Language Model for Financial Domain
Raj Shah | Kunal Chawla | Dheeraj Eidnani | Agam Shah | Wendi Du | Sudheer Chava | Natraj Raman | Charese Smiley | Jiaao Chen | Diyi Yang

Pre-trained language models have shown impressive performance on a variety of tasks and domains. Previous research on financial language models usually employs a generic training scheme to train standard model architectures, without completely leveraging the richness of the financial data. We propose a novel domain specific Financial LANGuage model (FLANG) which uses financial keywords and phrases for better masking, together with span boundary objective and in-filing objective. Additionally, the evaluation benchmarks in the field have been limited. To this end, we contribute the Financial Language Understanding Evaluation (FLUE), an open-source comprehensive suite of benchmarks for the financial domain. These include new benchmarks across 5 NLP tasks in financial domain as well as common benchmarks used in the previous research. Experiments on these benchmarks suggest that our model outperforms those in prior literature on a variety of NLP tasks. Our models, code and benchmark data will be made publicly available on Github and Huggingface.

Retrieval as Attention: End-to-end Learning of Retrieval and Reading within a Single Transformer
Zhengbao Jiang | Luyu Gao | Zhiruo Wang | Jun Araki | Haibo Ding | Jamie Callan | Graham Neubig

Systems for knowledge-intensive tasks such as open-domain question answering (QA) usually consist of two stages: efficient retrieval of relevant documents from a large corpus and detailed reading of the selected documents. This is usually done through two separate models, a retriever that encodes the query and finds nearest neighbors, and a reader based on Transformers. These two components are usually modeled separately, which necessitates a cumbersome implementation and is awkward to optimize in an end-to-end fashion. In this paper, we revisit this design and eschew the separate architecture and training in favor of a single Transformer that performs retrieval as attention (RAA), and end-to-end training solely based on supervision from the end QA task. We demonstrate for the first time that an end-to-end trained single Transformer can achieve both competitive retrieval and QA performance on in-domain datasets, matching or even slightly outperforming state-of-the-art dense retrievers and readers. Moreover, end-to-end adaptation of our model significantly boosts its performance on out-of-domain datasets in both supervised and unsupervised settings, making our model a simple and adaptable end-to-end solution for knowledge-intensive tasks.

Reproducibility in Computational Linguistics: Is Source Code Enough?
Mohammad Arvan | Luís Pina | Natalie Parde

The availability of source code has been put forward as one of the most critical factors for improving the reproducibility of scientific research. This work studies trends in source code availability at major computational linguistics conferences, namely, ACL, EMNLP, LREC, NAACL, and COLING. We observe positive trends, especially in conferences that actively promote reproducibility. We follow this by conducting a reproducibility study of eight papers published in EMNLP 2021, finding that source code releases leave much to be desired. Moving forward, we suggest all conferences require self-contained artifacts and provide a venue to evaluate such artifacts at the time of publication. Authors can include small-scale experiments and explicit scripts to generate each result to improve the reproducibility of their work.

Generating Information-Seeking Conversations from Unlabeled Documents
Gangwoo Kim | Sungdong Kim | Kang Min Yoo | Jaewoo Kang

Synthesizing datasets for conversational question answering (CQA) from unlabeled documents remains challenging due to its interactive nature.Moreover, while modeling information needs is an essential key, only few studies have discussed it.In this paper, we introduce a novel framework, **SimSeek**, (**Sim**ulating information-**Seek**ing conversation from unlabeled documents), and compare its two variants.In our baseline, **SimSeek-sym**, a questioner generates follow-up questions upon the predetermined answer by an answerer.On the contrary, **SimSeek-asym** first generates the question and then finds its corresponding answer under the conversational context.Our experiments show that they can synthesize effective training resources for CQA and conversational search tasks.As a result, conversations from **SimSeek-asym** not only make more improvements in our experiments but also are favorably reviewed in a human evaluation.We finally release a large-scale resource of synthetic conversations, **Wiki-SimSeek**, containing 2 million CQA pairs built upon Wikipedia documents.With the dataset, our CQA model achieves the state-of-the-art performance on a recent CQA benchmark, QuAC.The code and dataset are available at

Distill The Image to Nowhere: Inversion Knowledge Distillation for Multimodal Machine Translation
Ru Peng | Yawen Zeng | Jake Zhao

Past works on multimodal machine translation (MMT) elevate bilingual setup by incorporating additional aligned vision information.However, an image-must requirement of the multimodal dataset largely hinders MMT’s development — namely that it demands an aligned form of [image, source text, target text].This limitation is generally troublesome during the inference phase especially when the aligned image is not provided as in the normal NMT setup.Thus, in this work, we introduce IKD-MMT, a novel MMT framework to support the image-free inference phase via an inversion knowledge distillation scheme.In particular, a multimodal feature generator is executed with a knowledge distillation module, which directly generates the multimodal feature from (only) source texts as the input.While there have been a few prior works entertaining the possibility to support image-free inference for machine translation, their performances have yet to rival the image-must translation.In our experiments, we identify our method as the first image-free approach to comprehensively rival or even surpass (almost) all image-must frameworks, and achieved the state-of-the-art result on the often-used Multi30k benchmark. Our code and data are availableat:

A Multifaceted Framework to Evaluate Evasion, Content Preservation, and Misattribution in Authorship Obfuscation Techniques
Malik Altakrori | Thomas Scialom | Benjamin C. M. Fung | Jackie Chi Kit Cheung

Authorship obfuscation techniques have commonly been evaluated based on their ability to hide the author’s identity (evasion) while preserving the content of the original text. However, to avoid overstating the systems’ effectiveness, evasion detection must be evaluated using competitive identification techniques in settings that mimic real-life scenarios, and the outcomes of the content-preservation evaluation have to be interpretable by potential users of these obfuscation tools. Motivated by recent work on cross-topic authorship identification and content preservation in summarization, we re-evaluate different authorship obfuscation techniques on detection evasion and content preservation. Furthermore, we propose a new information-theoretic measure to characterize the misattribution harm that can be caused by detection evasion. Our results reveal key weaknesses in state-of-the-art obfuscation techniques and a surprisingly competitive effectiveness from a back-translation baseline in all evaluation aspects.

SafeText: A Benchmark for Exploring Physical Safety in Language Models
Sharon Levy | Emily Allaway | Melanie Subbiah | Lydia Chilton | Desmond Patton | Kathleen McKeown | William Yang Wang

Understanding what constitutes safe text is an important issue in natural language processing and can often prevent the deployment of models deemed harmful and unsafe. One such type of safety that has been scarcely studied is commonsense physical safety, i.e. text that is not explicitly violent and requires additional commonsense knowledge to comprehend that it leads to physical harm. We create the first benchmark dataset, SafeText, comprising real-life scenarios with paired safe and physically unsafe pieces of advice. We utilize SafeText to empirically study commonsense physical safety across various models designed for text generation and commonsense reasoning tasks. We find that state-of-the-art large language models are susceptible to the generation of unsafe text and have difficulty rejecting unsafe advice. As a result, we argue for further studies of safety and the assessment of commonsense physical safety in models before release.

Ground-Truth Labels Matter: A Deeper Look into Input-Label Demonstrations
Kang Min Yoo | Junyeob Kim | Hyuhng Joon Kim | Hyunsoo Cho | Hwiyeol Jo | Sang-Woo Lee | Sang-goo Lee | Taeuk Kim

Despite recent explosion of interests in in-context learning, the underlying mechanism and the precise impact of the quality of demonstrations remain elusive.Intuitively, ground-truth labels should have as much impact in in-context learning (ICL) as supervised learning, but recent work reported that the input-label correspondence is significantly less important than previously thought.Intrigued by this counter-intuitive observation, we re-examine the importance of ground-truth labels in in-context learning.With the introduction of two novel metrics, namely Label-Correctness Sensitivity and Ground-truth Label Effect Ratio (GLER), we were able to conduct quantifiable analysis on the impact of ground-truth label demonstrations.Through extensive analyses, we find that the correct input-label mappings can have varying impacts on the downstream in-context learning performances, depending on the experimental configuration.Through additional studies, we identify key components, such as the verbosity of prompt templates and the language model size, as the controlling factor to achieve more noise-resilient ICL.

D4: a Chinese Dialogue Dataset for Depression-Diagnosis-Oriented Chat
Binwei Yao | Chao Shi | Likai Zou | Lingfeng Dai | Mengyue Wu | Lu Chen | Zhen Wang | Kai Yu

In a depression-diagnosis-directed clinical session, doctors initiate a conversation with ample emotional support that guides the patients to expose their symptoms based on clinical diagnosis criteria. Such a dialogue system is distinguished from existing single-purpose human-machine dialog systems, as it combines task-oriented and chit-chats with uniqueness in dialogue topics and procedures. However, due to the social stigma associated with mental illness, the dialogue data related to depression consultation and diagnosis are rarely disclosed. Based on clinical depression diagnostic criteria ICD-11 and DSM-5, we designed a 3-phase procedure to construct D4: a Chinese Dialogue Dataset for Depression-Diagnosis-Oriented Chat, which simulates the dialogue between doctors and patients during the diagnosis of depression, including diagnosis results and symptom summary given by professional psychiatrists for each conversation. Upon the newly-constructed dataset, four tasks mirroring the depression diagnosis process are established: response generation, topic prediction, dialog summary, and severity classification of depressive episode and suicide risk. Multi-scale evaluation results demonstrate that a more empathy-driven and diagnostic-accurate consultation dialogue system trained on our dataset can be achieved compared to rule-based bots.

Exploiting domain-slot related keywords description for Few-Shot Cross-Domain Dialogue State Tracking
Gao Qixiang | Guanting Dong | Yutao Mou | Liwen Wang | Chen Zeng | Daichi Guo | Mingyang Sun | Weiran Xu

Collecting dialogue data with domain-slot-value labels for dialogue state tracking (DST) could be a costly process. In this paper, we propose a novel framework based on domain-slot related description to tackle the challenge of few-shot cross-domain DST. Specifically, we design an extraction module to extract domain-slot related verbs and nouns in the dialogue. Then, we integrates them into the description, which aims to prompt the model to identify the slot information. Furthermore, we introduce a random sampling strategy to improve the domain generalization ability of the model. We utilize a pre-trained model to encode contexts and description and generates answers with an auto-regressive manner. Experimental results show that our approaches substantially outperform the existing few-shot DST methods on MultiWOZ and gain strong improvements on the slot accuracy comparing to existing slot description methods.

CoCoa: An Encoder-Decoder Model for Controllable Code-switched Generation
Sneha Mondal | Ritika . | Shreya Pathak | Preethi Jyothi | Aravindan Raghuveer

Code-switching has seen growing interest in recent years as an important multilingual NLP phenomenon. Generating code-switched text for data augmentation has been sufficiently well-explored. However, there is no prior work on generating code-switched text with fine-grained control on the degree of code-switching and the lexical choices used to convey formality. We present CoCoa, an encoder-decoder translation model that converts monolingual Hindi text to Hindi-English code-switched text with both encoder-side and decoder-side interventions to achieve fine-grained controllable generation. CoCoa can be invoked at test-time to synthesize code-switched text that is simultaneously faithful to syntactic and lexical attributes relevant to code-switching. CoCoa outputs were subjected to rigorous subjective and objective evaluations. Human evaluations establish that our outputs are of superior quality while being faithful to desired attributes. We show significantly improved BLEU scores when compared with human-generated code-switched references. Compared to competitive baselines, we show 10% reduction in perplexity on a language modeling task and also demonstrate clear improvements on a downstream code-switched sentiment analysis task.

Towards Climate Awareness in NLP Research
Daniel Hershcovich | Nicolas Webersinke | Mathias Kraus | Julia Bingler | Markus Leippold

The climate impact of AI, and NLP research in particular, has become a serious issue given the enormous amount of energy that is increasingly being used for training and running computational models. Consequently, increasing focus is placed on efficient NLP. However, this important initiative lacks simple guidelines that would allow for systematic climate reporting of NLP research. We argue that this deficiency is one of the reasons why very few publications in NLP report key figures that would allow a more thorough examination of environmental impact, and present a quantitative survey to demonstrate this. As a remedy, we propose a climate performance model card with the primary purpose of being practically usable with only limited information about experiments and the underlying computer hardware. We describe why this step is essential to increase awareness about the environmental impact of NLP research and, thereby, paving the way for more thorough discussions.

Navigating Connected Memories with a Task-oriented Dialog System
Satwik Kottur | Seungwhan Moon | Alborz Geramifard | Babak Damavandi

Recent years have seen an increasing trend in the volume of personal media captured by users, thanks to the advent of smartphones and smart glasses, resulting in large media collections. Despite conversation being an intuitive human-computer interface, current efforts focus mostly on single-shot natural language based media retrieval to aid users query their media and re-live their memories. This severely limits the search functionality as users can neither ask follow-up queries nor obtain information without first formulating a single-turn query.In this work, we propose dialogs for connected memories as a powerful tool to empower users to search their media collection through a multi-turn, interactive conversation. Towards this, we collect a new task-oriented dialog dataset COMET, which contains 11.5k user↔assistant dialogs (totalling 103k utterances), grounded in simulated personal memory graphs. We employ a resource-efficient, two-phase data collection pipeline that uses: (1) a novel multimodal dialog simulator that generates synthetic dialog flows grounded in memory graphs, and, (2) manual paraphrasing to obtain natural language utterances. We analyze COMET, formulate four main tasks to benchmark meaningful progress, and adopt state-of-the-art language models as strong baselines, in order to highlight the multimodal challenges captured by our dataset.

Language Model Decomposition: Quantifying the Dependency and Correlation of Language Models
Hao Zhang

Pre-trained language models (LMs), such as BERT (Devlin et al., 2018) and its variants, have led to significant improvements on various NLP tasks in past years. However, a theoretical framework for studying their relationships is still missing. In this paper, we fill this gap by investigating the linear dependency between pre-trained LMs. The linear dependency of LMs is defined analogously to the linear dependency of vectors. We propose Language Model Decomposition (LMD) to represent a LM using a linear combination of other LMs as basis, and derive the closed-form solution. A goodness-of-fit metric for LMD similar to the coefficient of determination is defined and used to measure the linear dependency of a set of LMs. In experiments, we find that BERT and eleven (11) BERT-like LMs are 91% linearly dependent. This observation suggests that current state-of-the-art (SOTA) LMs are highly “correlated”. To further advance SOTA we need more diverse and novel LMs that are less dependent on existing LMs.

SynGEC: Syntax-Enhanced Grammatical Error Correction with a Tailored GEC-Oriented Parser
Yue Zhang | Bo Zhang | Zhenghua Li | Zuyi Bao | Chen Li | Min Zhang

This work proposes a syntax-enhanced grammatical error correction (GEC) approach named SynGEC that effectively incorporates dependency syntactic information into the encoder part of GEC models. The key challenge for this idea is that off-the-shelf parsers are unreliable when processing ungrammatical sentences. To confront this challenge, we propose to build a tailored GEC-oriented parser (GOPar) using parallel GEC training data as a pivot. First, we design an extended syntax representation scheme that allows us to represent both grammatical errors and syntax in a unified tree structure. Then, we obtain parse trees of the source incorrect sentences by projecting trees of the target correct sentences. Finally, we train GOPar with such projected trees. For GEC, we employ the graph convolution network to encode source-side syntactic information produced by GOPar, and fuse them with the outputs of the Transformer encoder. Experiments on mainstream English and Chinese GEC datasets show that our proposed SynGEC approach consistently and substantially outperforms strong baselines and achieves competitive performance. Our code and data are all publicly available at

Varifocal Question Generation for Fact-checking
Nedjma Ousidhoum | Zhangdie Yuan | Andreas Vlachos

Fact-checking requires retrieving evidence related to a claim under investigation. The task can be formulated as question generation based on a claim, followed by question answering.However, recent question generation approaches assume that the answer is known and typically contained in a passage given as input,whereas such passages are what is being sought when verifying a claim.In this paper, we present Varifocal, a method that generates questions based on different focal points within a given claim, i.e. different spans of the claim and its metadata, such as its source and date.Our method outperforms previous work on a fact-checking question generation dataset on a wide range of automatic evaluation metrics.These results are corroborated by our manual evaluation, which indicates that our method generates more relevant and informative questions.We further demonstrate the potential of focal points in generating sets of clarification questions for product descriptions.

Bilingual Lexicon Induction for Low-Resource Languages using Graph Matching via Optimal Transport
Kelly Marchisio | Ali Saad-Eldin | Kevin Duh | Carey Priebe | Philipp Koehn

Bilingual lexicons form a critical component of various natural language processing applications, including unsupervised and semisupervised machine translation and crosslingual information retrieval. In this work, we improve bilingual lexicon induction performance across 40 language pairs with a graph-matching method based on optimal transport. The method is especially strong with low amounts of supervision.

Whose Language Counts as High Quality? Measuring Language Ideologies in Text Data Selection
Suchin Gururangan | Dallas Card | Sarah Dreier | Emily Gade | Leroy Wang | Zeyu Wang | Luke Zettlemoyer | Noah A. Smith

Language models increasingly rely on massive web crawls for diverse text data. However, these sources are rife with undesirable content. As such, resources like Wikipedia, books, and news often serve as anchors for automatically selecting web text most suitable for language modeling, a process typically referred to as quality filtering. Using a new dataset of U.S. high school newspaper articles—written by students from across the country—we investigate whose language is preferred by the quality filter used for GPT-3. We find that newspapers from larger schools, located in wealthier, educated, and urban zones (ZIP codes) are more likely to be classified as high quality. We also show that this quality measurement is unaligned with other sensible metrics, such as factuality or literary acclaim. We argue that privileging any corpus as high quality entails a language ideology, and more care is needed to construct training corpora for language models, with better transparency and justification for the inclusion or exclusion of various texts.

ConReader: Exploring Implicit Relations in Contracts for Contract Clause Extraction
Weiwen Xu | Yang Deng | Wenqiang Lei | Wenlong Zhao | Tat-Seng Chua | Wai Lam

We study automatic Contract Clause Extraction (CCE) by modeling implicit relations in legal contracts. Existing CCE methods mostly treat contracts as plain text, creating a substantial barrier to understanding contracts of high complexity. In this work, we first comprehensively analyze the complexity issues of contracts and distill out three implicit relations commonly found in contracts, namely, 1) Long-range Context Relation that captures the correlations of distant clauses; 2) Term-Definition Relation that captures the relation between important terms with their corresponding definitions, and 3) Similar Clause Relation that captures the similarities between clauses of the same type. Then we propose a novel framework ConReader to exploit the above three relations for better contract understanding and improving CCE. Experimental results show that ConReader makes the prediction more interpretable and achieves new state-of-the-art on two CCE tasks in both conventional and zero-shot settings.

Training Dynamics for Curriculum Learning: A Study on Monolingual and Cross-lingual NLU
Fenia Christopoulou | Gerasimos Lampouras | Ignacio Iacobacci

Curriculum Learning (CL) is a technique of training models via ranking examples in a typically increasing difficulty trend with the aim of accelerating convergence and improving generalisability. Current approaches for Natural Language Understanding (NLU) tasks use CL to improve in-distribution data performance often via heuristic-oriented or task-agnostic difficulties. In this work, instead, we employ CL for NLU by taking advantage of training dynamics as difficulty metrics, i.e., statistics that measure the behavior of the model at hand on specific task-data instances during training and propose modifications of existing CL schedulers based on these statistics. Differently from existing works, we focus on evaluating models on in-distribution (ID), out-of-distribution (OOD) as well as zero-shot (ZS) cross-lingual transfer datasets. We show across several NLU tasks that CL with training dynamics can result in better performance mostly on zero-shot cross-lingual transfer and OOD settings with improvements up by 8.5% in certain cases. Overall, experiments indicate that training dynamics can lead to better performing models with smoother training compared to other difficulty metrics while being 20% faster on average. In addition, through analysis we shed light on the correlations of task-specific versus task-agnostic metrics.

Revisiting Parameter-Efficient Tuning: Are We Really There Yet?
Guanzheng Chen | Fangyu Liu | Zaiqiao Meng | Shangsong Liang

Parameter-Efficient Tuning (PETuning) methods have been deemed by many as the new paradigm for using pretrained language models (PLMs). By tuning just a fraction amount of parameters comparing to full model finetuning, PETuning methods claim to have achieved performance on par with or even better than finetuning. In this work, we take a step back and re-examine these PETuning methods by conducting the first comprehensive investigation into the training and evaluation of them. We found the problematic validation and testing practice in current studies, when accompanied by the instability nature of PETuning methods, has led to unreliable conclusions. When being compared under a truly fair evaluation protocol, PETuning cannot yield consistently competitive performance while finetuning remains to be the best-performing method in medium- and high-resource settings. We delve deeper into the cause of the instability and observed that the number of trainable parameters and training iterations are two main factors: reducing trainable parameters and prolonging training iterations may lead to higher stability in PETuning methods.

Transfer Learning from Semantic Role Labeling to Event Argument Extraction with Template-based Slot Querying
Zhisong Zhang | Emma Strubell | Eduard Hovy

In this work, we investigate transfer learning from semantic role labeling (SRL) to event argument extraction (EAE), considering their similar argument structures. We view the extraction task as a role querying problem, unifying various methods into a single framework. There are key discrepancies on role labels and distant arguments between semantic role and event argument annotations. To mitigate these discrepancies, we specify natural language-like queries to tackle the label mismatch problem and devise argument augmentation to recover distant arguments. We show that SRL annotations can serve as a valuable resource for EAE, and a template-based slot querying strategy is especially effective for facilitating the transfer. In extensive evaluations on two English EAE benchmarks, our proposed model obtains impressive zero-shot results by leveraging SRL annotations, reaching nearly 80% of the fullysupervised scores. It further provides benefits in low-resource cases, where few EAE annotations are available. Moreover, we show that our approach generalizes to cross-domain and multilingual scenarios.

Calibrating Zero-shot Cross-lingual (Un-)structured Predictions
Zhengping Jiang | Anqi Liu | Benjamin Van Durme

We investigate model calibration in the setting of zero-shot cross-lingual transfer with large-scale pre-trained language models. The level of model calibration is an important metric for evaluating the trustworthiness of predictive models. There exists an essential need for model calibration when natural language models are deployed in critical tasks. We study different post-training calibration methods in structured and unstructured prediction tasks. We find that models trained with data from the source language become less calibrated when applied to the target language and that calibration errors increase with intrinsic task difficulty and relative sparsity of training data. Moreover, we observe a potential connection between the level of calibration error and an earlier proposed measure of the distance from English to other languages. Finally, our comparison demonstrates that among other methods Temperature Scaling (TS) generalizes well to distant languages, but TS fails to calibrate more complex confidence estimation in structured predictions compared to more expressive alternatives like Gaussian Process Calibration.

PRINCE: Prefix-Masked Decoding for Knowledge Enhanced Sequence-to-Sequence Pre-Training
Song Xu | Haoran Li | Peng Yuan | Youzheng Wu | Xiaodong He

Pre-trained Language Models (PLMs) have shown effectiveness in various Natural Language Processing (NLP) tasks. Denoising autoencoder is one of the most successful pre-training frameworks, learning to recompose the original text given a noise-corrupted one. The existing studies mainly focus on injecting noises into the input. This paper introduces a simple yet effective pre-training paradigm, equipped with a knowledge-enhanced decoder that predicts the next entity token with noises in the prefix, explicitly strengthening the representation learning of entities that span over multiple input tokens. Specifically, when predicting the next token within an entity, we feed masks into the prefix in place of some of the previous ground-truth tokens that constitute the entity. Our model achieves new state-of-the-art results on two knowledge-driven data-to-text generation tasks with up to 2% BLEU gains.

How Far are We from Robust Long Abstractive Summarization?
Huan Yee Koh | Jiaxin Ju | He Zhang | Ming Liu | Shirui Pan

Abstractive summarization has made tremendous progress in recent years. In this work, we perform fine-grained human annotations to evaluate long document abstractive summarization systems (i.e., models and metrics) with the aim of implementing them to generate reliable summaries. For long document abstractive models, we show that the constant strive for state-of-the-art ROUGE results can lead us to generate more relevant summaries but not factual ones. For long document evaluation metrics, human evaluation results show that ROUGE remains the best at evaluating the relevancy of a summary. It also reveals important limitations of factuality metrics in detecting different types of factual errors and the reasons behind the effectiveness of BARTScore. We then suggest promising directions in the endeavor of developing factual consistency metrics. Finally, we release our annotated long document dataset with the hope that it can contribute to the development of metrics across a broader range of summarization settings.

Measuring Context-Word Biases in Lexical Semantic Datasets
Qianchu Liu | Diana McCarthy | Anna Korhonen

State-of-the-art pretrained contextualized models (PCM) eg. BERT use tasks such as WiC and WSD to evaluate their word-in-context representations. This inherently assumes that performance in these tasks reflect how well a model represents the coupled word and context semantics. We question this assumption by presenting the first quantitative analysis on the context-word interaction being tested in major contextual lexical semantic tasks. To achieve this, we run probing baselines on masked input, and propose measures to calculate and visualize the degree of context or word biases in existing datasets. The analysis was performed on both models and humans. Our findings demonstrate that models are usually not being tested for word-in-context semantics in the same way as humans are in these tasks, which helps us better understand the model-human gap. Specifically, to PCMs, most existing datasets fall into the extreme ends (the retrieval-based tasks exhibit strong target word bias while WiC-style tasks and WSD show strong context bias); In comparison, humans are less biased and achieve much better performance when both word and context are available than with masked input. We recommend our framework for understanding and controlling these biases for model interpretation and future task design.

Iteratively Prompt Pre-trained Language Models for Chain of Thought
Boshi Wang | Xiang Deng | Huan Sun

While Pre-trained Language Models (PLMs) internalize a great amount of world knowledge, they have been shown incapable of recalling these knowledge to solve tasks requiring complex & multi-step reasoning. Similar to how humans develop a “chain of thought” for these tasks, how can we equip PLMs with such abilities? In this work, we explore an iterative prompting framework, a new prompting paradigm which progressively elicits relevant knowledge from PLMs for multi-step inference. We identify key limitations of existing prompting methods, namely they are either restricted to queries with a single identifiable relation/predicate, or being agnostic to input contexts, which makes it difficult to capture variabilities across different inference steps. We propose an iterative context-aware prompter, which addresses these limitations by learning to dynamically synthesize prompts conditioned on the current step’s contexts. Experiments on three datasets involving multi-step reasoning show the effectiveness of the iterative scheme and the context-aware prompter design.

Unobserved Local Structures Make Compositional Generalization Hard
Ben Bogin | Shivanshu Gupta | Jonathan Berant

While recent work has shown that sequence-to-sequence models struggle to generalize to new compositions (termed compositional generalization), little is known on what makes compositional generalization hard on a particular test instance. In this work, we investigate the factors that make generalization to certain test instances challenging. We first substantiate that some examples are more difficult than others by showing that different models consistently fail or succeed on the same test instances. Then, we propose a criterion for the difficulty of an example: a test instance is hard if it contains a local structure that was not observed at training time. We formulate a simple decision rule based on this criterion and empirically show it predicts instance-level generalization well across 5 different semantic parsing datasets, substantially better than alternative decision rules. Last, we show local structures can be leveraged for creating difficult adversarial compositional splits and also to improve compositional generalization under limited training budgets by strategically selecting examples for the training set.

Mitigating Data Sparsity for Short Text Topic Modeling by Topic-Semantic Contrastive Learning
Xiaobao Wu | Anh Tuan Luu | Xinshuai Dong

To overcome the data sparsity issue in short text topic modeling, existing methods commonly rely on data augmentation or the data characteristic of short texts to introduce more word co-occurrence information. However, most of them do not make full use of the augmented data or the data characteristic: they insufficiently learn the relations among samples in data, leading to dissimilar topic distributions of semantically similar text pairs. To better address data sparsity, in this paper we propose a novel short text topic modeling framework, Topic-Semantic Contrastive Topic Model (TSCTM). To sufficiently model the relations among samples, we employ a new contrastive learning method with efficient positive and negative sampling strategies based on topic semantics. This contrastive learning method refines the representations, enriches the learning signals, and thus mitigates the sparsity issue. Extensive experimental results show that our TSCTM outperforms state-of-the-art baselines regardless of the data augmentation availability, producing high-quality topics and topic distributions.

Back to the Future: Bidirectional Information Decoupling Network for Multi-turn Dialogue Modeling
Yiyang Li | Hai Zhao | Zhuosheng Zhang

Multi-turn dialogue modeling as a challenging branch of natural language understanding (NLU), aims to build representations for machines to understand human dialogues, which provides a solid foundation for multiple downstream tasks. Recent studies of dialogue modeling commonly employ pre-trained language models (PrLMs) to encode the dialogue history as successive tokens, which is insufficient in capturing the temporal characteristics of dialogues. Therefore, we propose Bidirectional Information Decoupling Network (BiDeN) as a universal dialogue encoder, which explicitly incorporates both the past and future contexts and can be generalized to a wide range of dialogue-related tasks. Experimental results on datasets of different downstream tasks demonstrate the universality and effectiveness of our BiDeN.

Calibration Meets Explanation: A Simple and Effective Approach for Model Confidence Estimates
Dongfang Li | Baotian Hu | Qingcai Chen

Calibration strengthens the trustworthiness of black-box models by producing better accurate confidence estimates on given examples. However, little is known about if model explanations can help confidence calibration. Intuitively, humans look at important features attributions and decide whether the model is trustworthy. Similarly, the explanations may tell us when the model might know and when it does not. Inspired by this, we propose a method named CME that leverages model explanations to make the model less confident with non-inductive attributions. The idea is that when the model is not highly confident, it is difficult to identify strong indications of any class, and the tokens accordingly do not have high attribution scores for any class and vice versa. We conduct extensive experiments on six datasets with two popular pre-trained language models in the in-domain and out-of-domain settings. The results show that CME improves calibration performance in all settings. The expected calibration errors are further reduced when combined with temperature scaling. Our findings highlight that model explanations can help calibrate posterior estimates.

Non-Autoregressive Neural Machine Translation: A Call for Clarity
Robin Schmidt | Telmo Pires | Stephan Peitz | Jonas Lööf

Non-autoregressive approaches aim to improve the inference speed of translation models by only requiring a single forward pass to generate the output sequence instead of iteratively producing each predicted token. Consequently, their translation quality still tends to be inferior to their autoregressive counterparts due to several issues involving output token interdependence. In this work, we take a step back and revisit several techniques that have been proposed for improving non-autoregressive translation models and compare their combined translation quality and speed implications under third-party testing environments. We provide novel insights for establishing strong baselines using length prediction or CTC-based architecture variants and contribute standardized BLEU, chrF++, and TER scores using sacreBLEU on four translation tasks, which crucially have been missing as inconsistencies in the use of tokenized BLEU lead to deviations of up to 1.7 BLEU points. Our open-sourced code is integrated into fairseq for reproducibility.

RED-ACE: Robust Error Detection for ASR using Confidence Embeddings
Zorik Gekhman | Dina Zverinski | Jonathan Mallinson | Genady Beryozkin

ASR Error Detection (AED) models aim to post-process the output of Automatic Speech Recognition (ASR) systems, in order to detect transcription errors. Modern approaches usually use text-based input, comprised solely of the ASR transcription hypothesis, disregarding additional signals from the ASR model. Instead, we utilize the ASR system’s word-level confidence scores for improving AED performance. Specifically, we add an ASR Confidence Embedding (ACE) layer to the AED model’s encoder, allowing us to jointly encode the confidence scores and the transcribed text into a contextualized representation. Our experiments show the benefits of ASR confidence scores for AED, their complementary effect over the textual signal, as well as the effectiveness and robustness of ACE for combining these signals. To foster further research, we publish a novel AED dataset consisting of ASR outputs on the LibriSpeech corpus with annotated transcription errors.

Fast-R2D2: A Pretrained Recursive Neural Network based on Pruned CKY for Grammar Induction and Text Representation
Xiang Hu | Haitao Mi | Liang Li | Gerard de Melo

Chart-based models have shown great potential in unsupervised grammar induction, running recursively and hierarchically, but requiring O(n³) time-complexity. The Recursive Transformer based on Differentiable Trees (R2D2) makes it possible to scale to large language model pretraining even with a complex tree encoder, by introducing a heuristic pruning method.However, its rule-based pruning process suffers from local optima and slow inference. In this paper, we propose a unified R2D2 method that overcomes these issues. We use a top-down unsupervised parser as a model-guided pruning method, which also enables parallel encoding during inference. Our parser casts parsing as a split point scoring task by first scoring all split points for a given sentence and then using the highest-scoring one to recursively split a span into two parts. The reverse order of the splits is considered as the order of pruning in the encoder. We optimize the unsupervised parser by minimizing the Kullback–Leibler distance between tree probabilities from the parser and the R2D2 model.Our experiments show that our Fast-R2D2 significantly improves the grammar induction quality and achieves competitive results in downstream tasks.

A Localized Geometric Method to Match Knowledge in Low-dimensional Hyperbolic Space
Bo Hui | Tian Xia | Wei-Shinn Ku

Matching equivalent entities across Knowledge graphs is a pivotal step for knowledge fusion. Previous approaches usually study the problem in Euclidean space. However, recent works have shown that hyperbolic space has a higher capacity than Euclidean space and hyperbolic embedding can represent the hierarchical structure in a knowledge graph. In this paper, we propose a localized geometric method to find equivalent entities in hyperbolic space. Specifically, we use a hyperbolic neural network to encode the lingual information of entities and the structure of both knowledge graphs into a low-dimensional hyperbolic space. To address the asymmetry of structure on different KGs and the localized nature of relations, we learn an instance-specific geometric mapping function based on rotation to match entity pairs. A contrastive loss function is used to train the model. The experiment verifies the power of low-dimensional hyperbolic space for entity matching and shows that our method outperforms the state of the art by a large margin.

Memory-assisted prompt editing to improve GPT-3 after deployment
Aman Madaan | Niket Tandon | Peter Clark | Yiming Yang

Large LMs such as GPT-3 are powerful, but can commit mistakes that are obvious to humans. For example, GPT-3 would mistakenly interpret “What word is similar to good?” to mean a homophone, while the user intended a synonym. Our goal is to effectively correct such errors via user interactions with the system but without retraining, which will be prohibitively costly. We pair GPT-3 with a growing memory of recorded cases where the model misunderstood the user’s intents, along with user feedback for clarification. Such a memory allows our system to produce enhanced prompts for any new query based on the user feedback for error correction on similar cases in the past. On four tasks (two lexical tasks, two advanced ethical reasoning tasks), we show how a (simulated) user can interactively teach a deployed GPT-3, substantially increasing its accuracy over the queries with different kinds of misunderstandings by the GPT-3. Our approach is a step towards the low-cost utility enhancement for very large pre-trained LMs.

LVP-M3: Language-aware Visual Prompt for Multilingual Multimodal Machine Translation
Hongcheng Guo | Jiaheng Liu | Haoyang Huang | Jian Yang | Zhoujun Li | Dongdong Zhang | Zheng Cui

Multimodal Machine Translation (MMT) focuses on enhancing text-only translation with visual features, which has attracted considerable attention from both natural language processing and computer vision communities. Recent advances still struggle to train a separate model for each language pair, which is costly and unaffordable when the number of languages increases in the real world. In other words, the multilingual multimodal machine translation (Multilingual MMT) task has not been investigated, which aims to handle the aforementioned issues by providing a shared semantic space for multiple languages. Besides, the image modality has no language boundaries, which is superior to bridging the semantic gap between languages. To this end,we first propose the Multilingual MMT task by establishing two new Multilingual MMT benchmark datasets covering seven languages.Then, an effective baseline LVP-M3 using visual prompts is proposed to support translations between different languages,which includes three stages (token encoding, language-aware visual prompt generation, and language translation). Extensive experimental results on our constructed benchmark datasets demonstrate the effectiveness of LVP-M3 method for Multilingual MMT.

PromptEHR: Conditional Electronic Healthcare Records Generation with Prompt Learning
Zifeng Wang | Jimeng Sun

Accessing longitudinal multimodal Electronic Healthcare Records (EHRs) is challenging due to privacy concerns, which hinders the use of ML for healthcare applications. Synthetic EHRs generation bypasses the need to share sensitive real patient records. However, existing methods generate single-modal EHRs by unconditional generation or by longitudinal inference, which falls short of low flexibility and makes unrealistic EHRs. In this work, we propose to formulate EHRs generation as a text-to-text translation task by language models (LMs), which suffices to highly flexible event imputation during generation. We also design prompt learning to control the generation conditioned by numerical and categorical demographic features. We evaluate synthetic EHRs quality by two perplexity measures accounting for their longitudinal pattern (longitudinal imputation perplexity, lpl) and the connections cross modalities (cross-modality imputation perplexity, mpl). Moreover, we utilize two adversaries: membership and attribute inference attacks for privacy-preserving evaluation. Experiments on MIMIC-III data demonstrate the superiority of our methods on realistic EHRs generation (53.1% decrease of lpl and 45.3% decrease of mpl on average compared to the best baselines) with low privacy risks. Software is available at

ROSE: Robust Selective Fine-tuning for Pre-trained Language Models
Lan Jiang | Hao Zhou | Yankai Lin | Peng Li | Jie Zhou | Rui Jiang

Even though the large-scale language models have achieved excellent performances, they suffer from various adversarial attacks.A large body of defense methods has been proposed. However, they are still limited due to redundant attack search spaces and the inability to defend against various types of attacks.In this work, we present a novel fine-tuning approach called RObust SEletive fine-tuning (ROSE) to address this issue.ROSE conducts selective updates when adapting pre-trained models to downstream tasks, filtering out invaluable and unrobust updates of parameters.Specifically, we propose two strategies: the first-order and second-order ROSE for selecting target robust parameters.The experimental results show that ROSE achieves significant improvements in adversarial robustness on various downstream NLP tasks, and the ensemble method even surpasses both variants above.Furthermore, ROSE can be easily incorporated into existing fine-tuning methods to improve their adversarial robustness further.The empirical analysis confirms that ROSE eliminates unrobust spurious updates during fine-tuning, leading to solutions corresponding to flatter and wider optima than the conventional method.Code is available at

CodeRetriever: A Large Scale Contrastive Pre-Training Method for Code Search
Xiaonan Li | Yeyun Gong | Yelong Shen | Xipeng Qiu | Hang Zhang | Bolun Yao | Weizhen Qi | Daxin Jiang | Weizhu Chen | Nan Duan

In this paper, we propose the CodeRetriever model, which learns the function-level code semantic representations through large-scale code-text contrastive pre-training. We adopt two contrastive learning schemes in CodeRetriever: unimodal contrastive learning and bimodal contrastive learning. For unimodal contrastive learning, we design an unsupervised learning approach to build semantic-related code pairs based on the documentation and function name. For bimodal contrastive learning, we leverage the documentation and in-line comments of code to build code-text pairs. Both contrastive objectives can fully leverage large-scale code corpus for pre-training. Extensive experimental results show that CodeRetriever achieves new state-of-the-art with significant improvement over existing code pre-trained models, on eleven domain/language-specific code search tasks with six programming languages in different code granularity (function-level, snippet-level and statement-level).These results demonstrate the effectiveness and robustness of CodeRetriever.The codes and resources are available at

Open-Topic False Information Detection on Social Networks with Contrastive Adversarial Learning
Guanghui Ma | Chunming Hu | Ling Ge | Hong Zhang

Current works about false information detection based on conversation graphs on social networks focus primarily on two research streams from the standpoint of topic distribution: in-topic and cross-topic techniques, which assume that the data topic distribution is identical or cross, respectively. This signifies that all test data topics are seen or unseen by the model.However, these assumptions are too harsh for actual social networks that contain both seen and unseen topics simultaneously, hence restricting their practical application.In light of this, this paper develops a novel open-topic scenario that is better suited to actual social networks. In this open-topic scenario, we empirically find that the existing models suffer from impairment in the detection performance for seen or unseen topic data, resulting in poor overall model performance. To address this issue, we propose a novel Contrastive Adversarial Learning Network, CALN, that employs an unsupervised topic clustering method to capture topic-specific features to enhance the model’s performance for seen topics and an unsupervised adversarial learning method to align data representation distributions to enhance the model’s generalisation to unseen topics.Experiments on two benchmark datasets and a variety of graph neural networks demonstrate the effectiveness of our approach.

Mitigating Inconsistencies in Multimodal Sentiment Analysis under Uncertain Missing Modalities
Jiandian Zeng | Jiantao Zhou | Tianyi Liu

For the missing modality problem in Multimodal Sentiment Analysis (MSA), the inconsistency phenomenon occurs when the sentiment changes due to the absence of a modality. The absent modality that determines the overall semantic can be considered as a key missing modality. However, previous works all ignored the inconsistency phenomenon, simply discarding missing modalities or solely generating associated features from available modalities. The neglect of the key missing modality case may lead to incorrect semantic results. To tackle the issue, we propose an Ensemble-based Missing Modality Reconstruction (EMMR) network to detect and recover semantic features of the key missing modality. Specifically, we first learn joint representations with remaining modalities via a backbone encoder-decoder network. Then, based on the recovered features, we check the semantic consistency to determine whether the absent modality is crucial to the overall sentiment polarity. Once the inconsistency problem due to the key missing modality exists, we integrate several encoder-decoder approaches for better decision making. Extensive experiments and analyses are conducted on CMU-MOSI and IEMOCAP datasets, validating the superiority of the proposed method.

ConvTrans: Transforming Web Search Sessions for Conversational Dense Retrieval
Kelong Mao | Zhicheng Dou | Hongjin Qian | Fengran Mo | Xiaohua Cheng | Zhao Cao

Conversational search provides users with a natural and convenient new search experience. Recently, conversational dense retrieval has shown to be a promising technique for realizing conversational search. However, as conversational search systems have not been widely deployed, it is hard to get large-scale real conversational search sessions and relevance labels to support the training of conversational dense retrieval. To tackle this data scarcity problem, previous methods focus on developing better few-shot learning approaches or generating pseudo relevance labels, but the data they use for training still heavily rely on manual generation.In this paper, we present ConvTrans, a data augmentation method that can automatically transform easily-accessible web search sessions into conversational search sessions to fundamentally alleviate the data scarcity problem for conversational dense retrieval. ConvTrans eliminates the gaps between these two types of sessions in terms of session quality and query form to achieve effective session transformation. Extensive evaluations on two widely used conversational search benchmarks, i.e., CAsT-19 and CAsT-20, demonstrate that the same model trained on the data generated by ConvTrans can achieve comparable retrieval performance as it trained on high-quality but expensive artificial conversational search data.

MUSIED: A Benchmark for Event Detection from Multi-Source Heterogeneous Informal Texts
Xiangyu Xi | Jianwei Lv | Shuaipeng Liu | Wei Ye | Fan Yang | Guanglu Wan

Event detection (ED) identifies and classifies event triggers from unstructured texts, serving as a fundamental task for information extraction. Despite the remarkable progress achieved in the past several years, most research efforts focus on detecting events from formal texts (e.g., news articles, Wikipedia documents, financial announcements). Moreover, the texts in each dataset are either from a single source or multiple yet relatively homogeneous sources. With massive amounts of user-generated text accumulating on the Web and inside enterprises, identifying meaningful events in these informal texts, usually from multiple heterogeneous sources, has become a problem of significant practical value. As a pioneering exploration that expands event detection to the scenarios involving informal and heterogeneous texts, we propose a new large-scale Chinese event detection dataset based on user reviews, text conversations, and phone conversations in a leading e-commerce platform for food service. We carefully investigate the proposed dataset’s textual informality and multi-domain heterogeneity characteristics by inspecting data samples quantitatively and qualitatively. Extensive experiments with state-of-the-art event detection methods verify the unique challenges posed by these characteristics, indicating that multi-domain informal event detection remains an open problem and requires further efforts. Our benchmark and code are released at

Reproducibility Issues for BERT-based Evaluation Metrics
Yanran Chen | Jonas Belouadi | Steffen Eger

Reproducibility is of utmost concern in machine learning and natural language processing (NLP). In the field of natural language generation (especially machine translation), the seminal paper of Post (2018) has pointed out problems of reproducibility of the dominant metric, BLEU, at the time of publication. Nowadays, BERT-based evaluation metrics considerably outperform BLEU. In this paper, we ask whether results and claims from four recent BERT-based metrics can be reproduced. We find that reproduction of claims and results often fails because of (i) heavy undocumented preprocessing involved in the metrics, (ii) missing code and (iii) reporting weaker results for the baseline metrics. (iv) In one case, the problem stems from correlating not to human scores but to a wrong column in the csv file, inflating scores by 5 points. Motivated by the impact of preprocessing, we then conduct a second study where we examine its effects more closely (for one of the metrics). We find that preprocessing can have large effects, especially for highly inflectional languages. In this case, the effect of preprocessing may be larger than the effect of the aggregation mechanism (e.g., greedy alignment vs. Word Mover Distance).

Improving Multi-task Stance Detection with Multi-task Interaction Network
Heyan Chai | Siyu Tang | Jinhao Cui | Ye Ding | Binxing Fang | Qing Liao

Stance detection aims to identify people’s standpoints expressed in the text towards a target, which can provide powerful information for various downstream tasks.Recent studies have proposed multi-task learning models that introduce sentiment information to boost stance detection.However, they neglect to explore capturing the fine-grained task-specific interaction between stance detection and sentiment tasks, thus degrading performance.To address this issue, this paper proposes a novel multi-task interaction network (MTIN) for improving the performance of stance detection and sentiment analysis tasks simultaneously.Specifically, we construct heterogeneous task-related graphs to automatically identify and adapt the roles that a word plays with respect to a specific task. Also, a multi-task interaction module is designed to capture the word-level interaction between tasks, so as to obtain richer task representations.Extensive experiments on two real-world datasets show that our proposed approach outperforms state-of-the-art methods in both stance detection and sentiment analysis tasks.

Neural-based Mixture Probabilistic Query Embedding for Answering FOL queries on Knowledge Graphs
Xiao Long | Liansheng Zhuang | Li Aodi | Shafei Wang | Houqiang Li

Query embedding (QE)—which aims to embed entities and first-order logical (FOL) queries in a vector space, has shown great power in answering FOL queries on knowledge graphs (KGs). Existing QE methods divide a complex query into a sequence of mini-queries according to its computation graph and perform logical operations on the answer sets of mini-queries to get answers. However, most of them assume that answer sets satisfy an individual distribution (e.g., Uniform, Beta, or Gaussian), which is often violated in real applications and limit their performance. In this paper, we propose a Neural-based Mixture Probabilistic Query Embedding Model (NMP-QEM) that encodes the answer set of each mini-query as a mixed Gaussian distribution with multiple means and covariance parameters, which can approximate any random distribution arbitrarily well in real KGs. Additionally, to overcome the difficulty in defining the closed solution of negation operation, we introduce neural-based logical operators of projection, intersection and negation for a mixed Gaussian distribution to answer all the FOL queries. Extensive experiments demonstrate that NMP-QEM significantly outperforms existing state-of-the-art methods on benchmark datasets. In NELL995, NMP-QEM achieves a 31% relative improvement over the state-of-the-art.

Improving Multi-turn Emotional Support Dialogue Generation with Lookahead Strategy Planning
Yi Cheng | Wenge Liu | Wenjie Li | Jiashuo Wang | Ruihui Zhao | Bang Liu | Xiaodan Liang | Yefeng Zheng

Providing Emotional Support (ES) to soothe people in emotional distress is an essential capability in social interactions. Most existing researches on building ES conversation systems only considered single-turn interactions with users, which was over-simplified. In comparison, multi-turn ES conversation systems can provide ES more effectively, but face several new technical challenges, including: (1) how to adopt appropriate support strategies to achieve the long-term dialogue goal of comforting the user’s emotion; (2) how to dynamically model the user’s state. In this paper, we propose a novel system MultiESC to address these issues. For strategy planning, drawing inspiration from the A* search algorithm, we propose lookahead heuristics to estimate the future user feedback after using particular strategies, which helps to select strategies that can lead to the best long-term effects. For user state modeling, MultiESC focuses on capturing users’ subtle emotional expressions and understanding their emotion causes. Extensive experiments show that MultiESC significantly outperforms competitive baselines in both dialogue generation and strategy planning.

Conformal Predictor for Improving Zero-Shot Text Classification Efficiency
Prafulla Kumar Choubey | Yu Bai | Chien-Sheng Wu | Wenhao Liu | Nazneen Rajani

Pre-trained language models (PLMs) have been shown effective for zero-shot (0shot) text classification. 0shot models based on natural language inference (NLI) and next sentence prediction (NSP) employ cross-encoder architecture and infer by making a forward pass through the model for each label-text pair separately. This increases the computational cost to make inferences linearly in the number of labels. In this work, we improve the efficiency of such cross-encoder-based 0shot models by restricting the number of likely labels using another fast base classifier-based conformal predictor (CP) calibrated on samples labeled by the 0shot model. Since a CP generates prediction sets with coverage guarantees, it reduces the number of target labels without excluding the most probable label based on the 0shot model. We experiment with three intent and two topic classification datasets. With a suitable CP for each dataset, we reduce the average inference time for NLI- and NSP-based models by 25.6% and 22.2% respectively, without dropping performance below the predefined error rate of 1%.

Effective and Efficient Query-aware Snippet Extraction for Web Search
Jingwei Yi | Fangzhao Wu | Chuhan Wu | Xiaolong Huang | Binxing Jiao | Guangzhong Sun | Xing Xie

Query-aware webpage snippet extraction is widely used in search engines to help users better understand the content of the returned webpages before clicking. The extracted snippet is expected to summarize the webpage in the context of the input query. Existing snippet extraction methods mainly rely on handcrafted features of overlapping words, which cannot capture deep semantic relationships between the query and webpages. Another idea is to extract the sentences which are most relevant to queries as snippets with existing text matching methods. However, these methods ignore the contextual information of webpages, which may be sub-optimal. In this paper, we propose an effective query-aware webpage snippet extraction method named DeepQSE. In DeepQSE, the concatenation of title, query and each candidate sentence serves as an input of query-aware sentence encoder, aiming to capture the fine-grained relevance between the query and sentences. Then, these query-aware sentence representations are modeled jointly through a document-aware relevance encoder to capture contextual information of the webpage. Since the query and each sentence are jointly modeled in DeepQSE, its online inference may be slow. Thus, we further propose an efficient version of DeepQSE, named Efficient-DeepQSE, which can significantly improve the inference speed of DeepQSE without affecting its performance. The core idea of Efficient-DeepQSE is to decompose the query-aware snippet extraction task into two stages, i.e., a coarse-grained candidate sentence selection stage where sentence representations can be cached, and a fine-grained relevance modeling stage. Experiments on two datasets validate the effectiveness and efficiency of our methods.

You Only Need One Model for Open-domain Question Answering
Haejun Lee | Akhil Kedia | Jongwon Lee | Ashwin Paranjape | Christopher Manning | Kyoung-Gu Woo

Recent approaches to Open-domain Question Answering refer to an external knowledge base using a retriever model, optionally rerank passages with a separate reranker model and generate an answer using another reader model. Despite performing related tasks, the models have separate parameters and are weakly-coupled during training. We propose casting the retriever and the reranker as internal passage-wise attention mechanisms applied sequentially within the transformer architecture and feeding computed representations to the reader, with the hidden representations progressively refined at each stage. This allows us to use a single question answering model trained end-to-end, which is a more efficient use of model capacity and also leads to better gradient flow. We present a pre-training method to effectively train this architecture and evaluate our model on the Natural Questions and TriviaQA open datasets. For a fixed parameter budget, our model outperforms the previous state-of-the-art model by 1.0 and 0.7 exact match scores.

Generative Entity Typing with Curriculum Learning
Siyu Yuan | Deqing Yang | Jiaqing Liang | Zhixu Li | Jinxi Liu | Jingyue Huang | Yanghua Xiao

Entity typing aims to assign types to the entity mentions in given texts. The traditional classification-based entity typing paradigm has two unignorable drawbacks: 1) it fails to assign an entity to the types beyond the predefined type set, and 2) it can hardly handle few-shot and zero-shot situations where many long-tail types only have few or even no training instances. To overcome these drawbacks, we propose a novel generative entity typing (GET) paradigm: given a text with an entity mention, the multiple types for the role that the entity plays in the text are generated with a pre-trained language model (PLM). However, PLMs tend to generate coarse-grained types after fine-tuning upon the entity typing dataset. In addition, only the heterogeneous training data consisting of a small portion of human-annotated data and a large portion of auto-generated but low-quality data are provided for model training. To tackle these problems, we employ curriculum learning (CL) to train our GET model on heterogeneous data, where the curriculum could be self-adjusted with the self-paced learning according to its comprehension of the type granularity and data heterogeneity. Our extensive experiments upon the datasets of different languages and downstream tasks justify the superiority of our GET model over the state-of-the-art entity typing models. The code has been released on

SetGNER: General Named Entity Recognition as Entity Set Generation
Yuxin He | Buzhou Tang

Recently, joint recognition of flat, nested and discontinuous entities has received increasing attention. Motivated by the observation that the target output of NER is essentially a set of sequences, we propose a novel entity set generation framework for general NER scenes in this paper. Different from sequence-to-sequence NER methods, our method does not force the entities to be generated in a predefined order and can get rid of the problem of error propagation and inefficient decoding. Distinguished from the set-prediction NER framework, our method treats each entity as a sequence and is capable of recognizing discontinuous mentions. Given an input sentence, the model first encodes the sentence in word-level and detects potential entity mentions based on the encoder’s output, then reconstructs entity mentions from the detected entity heads in parallel. To let the encoder of our model capture better right-to-left semantic structure, we also propose an auxiliary Inverse Generation Training task. Extensive experiments show that our model (w/o. Inverse Generation Training) outperforms state-of-the-art generative NER models by a large margin on two discontinuous NER datasets, two nested NER datasets and one flat NER dataset. Besides, the auxiliary Inverse Generation Training task is found to further improve the model’s performance on the five datasets.

Opinion Summarization by Weak-Supervision from Mix-structured Data
Yizhu Liu | Qi Jia | Kenny Zhu

Opinion summarization of multiple reviews suffers from the lack of reference summaries for training.Most previous approaches construct multiple reviews and their summary based on textual similarities between reviews,resulting in information mismatch between the review input and the summary. In this paper, we convert each review into a mixof structured and unstructured data, which we call opinion-aspect pairs (OAs) and implicit sentences (ISs).We propose a new method to synthesize training pairs of such mix-structured data as input and the textual summary as output,and design a summarization model with OA encoder and IS encoder.Experiments show that our approach outperforms previous methods on Yelp, Amazon and RottenTomatos datasets.

Multi-level Distillation of Semantic Knowledge for Pre-training Multilingual Language Model
Mingqi Li | Fei Ding | Dan Zhang | Long Cheng | Hongxin Hu | Feng Luo

Pre-trained multilingual language models play an important role in cross-lingual natural language understanding tasks. However, existing methods did not focus on learning the semantic structure of representation, and thus could not optimize their performance. In this paper, we propose Multi-level Multilingual Knowledge Distillation (MMKD), a novel method for improving multilingual language models. Specifically, we employ a teacher-student framework to adopt rich semantic representation knowledge in English BERT. We propose token-, word-, sentence-, and structure-level alignment objectives to encourage multiple levels of consistency between source-target pairs and correlation similarity between teacher and student models. We conduct experiments on cross-lingual evaluation benchmarks including XNLI, PAWS-X, and XQuAD. Experimental results show that MMKD outperforms other baseline models of similar size on XNLI and XQuAD and obtains comparable performance on PAWS-X. Especially, MMKD obtains significant performance gains on low-resource languages.

Empowering Dual-Encoder with Query Generator for Cross-Lingual Dense Retrieval
Houxing Ren | Linjun Shou | Ning Wu | Ming Gong | Daxin Jiang

In monolingual dense retrieval, lots of works focus on how to distill knowledge from cross-encoder re-ranker to dual-encoder retriever and these methods achieve better performance due to the effectiveness of cross-encoder re-ranker. However, we find that the performance of the cross-encoder re-ranker is heavily influenced by the number of training samples and the quality of negative samples, which is hard to obtain in the cross-lingual setting. In this paper, we propose to use a query generator as the teacher in the cross-lingual setting, which is less dependent on enough training samples and high-quality negative samples. In addition to traditional knowledge distillation, we further propose a novel enhancement method, which uses the query generator to help the dual-encoder align queries from different languages, but does not need any additional parallel sentences. The experimental results show that our method outperforms the state-of-the-art methods on two benchmark datasets.

R2F: A General Retrieval, Reading and Fusion Framework for Document-level Natural Language Inference
Hao Wang | Yixin Cao | Yangguang Li | Zhen Huang | Kun Wang | Jing Shao

Document-level natural language inference (DOCNLI) is a new challenging task in natural language processing, aiming at judging the entailment relationship between a pair of hypothesis and premise documents. Current datasets and baselines largely follow sentence-level settings, but fail to address the issues raised by longer documents. In this paper, we establish a general solution, named Retrieval, Reading and Fusion (R2F) framework, and a new setting, by analyzing the main challenges of DOCNLI: interpretability, long-range dependency, and cross-sentence inference. The basic idea of the framework is to simplify document-level task into a set of sentence-level tasks, and improve both performance and interpretability with the power of evidence. For each hypothesis sentence, the framework retrieves evidence sentences from the premise, and reads to estimate its credibility. Then the sentence-level results are fused to judge the relationship between the documents. For the setting, we contribute complementary evidence and entailment label annotation on hypothesis sentences, for interpretability study. Our experimental results show that R2F framework can obtain state-of-the-art performance and is robust for diverse evidence retrieval methods. Moreover, it can give more interpretable prediction results. Our model and code are released at

Revisiting Pre-trained Language Models and their Evaluation for Arabic Natural Language Processing
Abbas Ghaddar | Yimeng Wu | Sunyam Bagga | Ahmad Rashid | Khalil Bibi | Mehdi Rezagholizadeh | Chao Xing | Yasheng Wang | Xinyu Duan | Zhefeng Wang | Baoxing Huai | Xin Jiang | Qun Liu | Phillippe Langlais

There is a growing body of work in recent years to develop pre-trained language models (PLMs) for the Arabic language. This work addresses two major problems in existing Arabic PLMs that limit the progress of the Arabic NLU and NLG fields. First, existing Arabic PLMs are not well-explored and their pre-training can be improved significantly using a more methodical approach. Second, there is a lack of systematic and reproducible evaluation of these models in the literature. We revisit both the pre-training and evaluation of Arabic PLMs. In terms of pre-training, we explore the impact of the quality of the pretraining data, the size of the model, and the incorporation of character-level information on Arabic PLM. As a result, we release three new Arabic BERT-style models ( JABER, Char-JABER, and SABER), and two T5-style models (AT5S and AT5B). In terms of evaluation, we conduct a comprehensive empirical study to systematically evaluate the performance of existing state-of-the-art models on ALUE, a leaderboard-powered benchmark for Arabic NLU tasks, and on a subset of the Arabic generative tasks. We show that our models significantly outperform existing Arabic PLMs and achieve a new state-of-the-art performance on discriminative and generative Arabic NLU and NLG tasks. Our models and source code to reproduce results will be made available upon acceptance.

KECP: Knowledge Enhanced Contrastive Prompting for Few-shot Extractive Question Answering
Jianing Wang | Chengyu Wang | Minghui Qiu | Qiuhui Shi | Hongbin Wang | Jun Huang | Ming Gao

Extractive Question Answering (EQA) is one of the most essential tasks in Machine Reading Comprehension (MRC), which can be solved by fine-tuning the span selecting heads of Pre-trained Language Models (PLMs). However, most existing approaches for MRC may perform poorly in the few-shot learning scenario. To solve this issue, we propose a novel framework named Knowledge Enhanced Contrastive Prompt-tuning (KECP). Instead of adding pointer heads to PLMs, we introduce a seminal paradigm for EQA that transforms the task into a non-autoregressive Masked Language Modeling (MLM) generation problem. Simultaneously, rich semantics from the external knowledge base (KB) and the passage context support enhancing the query’s representations. In addition, to boost the performance of PLMs, we jointly train the model by the MLM and contrastive learning objectives. Experiments on multiple benchmarks demonstrate that our method consistently outperforms state-of-the-art approaches in few-shot settings by a large margin.

Knowledge Prompting in Pre-trained Language Model for Natural Language Understanding
Jianing Wang | Wenkang Huang | Minghui Qiu | Qiuhui Shi | Hongbin Wang | Xiang Li | Ming Gao

Knowledge-enhanced Pre-trained Language Model (PLM) has recently received significant attention, which aims to incorporate factual knowledge into PLMs. However, most existing methods modify the internal structures of fixed types of PLMs by stacking complicated modules, and introduce redundant and irrelevant factual knowledge from knowledge bases (KBs). In this paper, to address these problems, we introduce a seminal knowledge prompting paradigm and further propose a knowledge-prompting-based PLM framework KP-PLM. This framework can be flexibly combined with existing mainstream PLMs. Specifically, we first construct a knowledge sub-graph from KBs for each context. Then we design multiple continuous prompts rules and transform the knowledge sub-graph into natural language prompts. To further leverage the factual knowledge from these prompts, we propose two novel knowledge-aware self-supervised tasks including prompt relevance inspection and masked prompt modeling. Extensive experiments on multiple natural language understanding (NLU) tasks show the superiority of KP-PLM over other state-of-the-art methods in both full-resource and low-resource settings. Our source codes will be released upon the acceptance of the paper.

On the Evaluation Metrics for Paraphrase Generation
Lingfeng Shen | Lemao Liu | Haiyun Jiang | Shuming Shi

In this paper we revisit automatic metrics for paraphrase evaluation and obtain two findings that disobey conventional wisdom: (1) Reference-free metrics achieve better performance than their reference-based counterparts. (2) Most commonly used metrics do not align well with human annotation.Underlying reasons behind the above findings are explored through additional experiments and in-depth analyses.Based on the experiments and analyses, we propose ParaScore, a new evaluation metric for paraphrase generation. It possesses the merits of reference-based and reference-free metrics and explicitly models lexical divergence. Based on our analysis and improvements, our proposed reference-based outperforms than reference-free metrics.Experimental results demonstrate that ParaScore significantly outperforms existing metrics.

Curriculum Learning Meets Weakly Supervised Multimodal Correlation Learning
Sijie Mai | Ya Sun | Haifeng Hu

In the field of multimodal sentiment analysis (MSA), a few studies have leveraged the inherent modality correlation information stored in samples for self-supervised learning. However, they feed the training pairs in a random order without consideration of difficulty. Without human annotation, the generated training pairs of self-supervised learning often contain noise. If noisy or hard pairs are used for training at the easy stage, the model might be stuck in bad local optimum. In this paper, we inject curriculum learning into weakly supervised multimodal correlation learning. The weakly supervised correlation learning leverages the label information to generate scores for negative pairs to learn a more discriminative embedding space, where negative pairs are defined as two unimodal embeddings from different samples. To assist the correlation learning, we feed the training pairs to the model according to difficulty by the proposed curriculum learning, which consists of elaborately designed scoring and feeding functions. The scoring function computes the difficulty of pairs using pre-trained and current correlation predictors, where the pairs with large losses are defined as hard pairs. Notably, the hardest pairs are discarded in our algorithm, which are assumed as noisy pairs. Moreover, the feeding function takes the difference of correlation losses as feedback to determine the feeding actions (‘stay’, ‘step back’, or ‘step forward’). The proposed method reaches state-of-the-art performance on MSA.

Rethinking Positional Encoding in Tree Transformer for Code Representation
Han Peng | Ge Li | Yunfei Zhao | Zhi Jin

Transformers are now widely used in code representation, and several recent works further develop tree Transformers to capture the syntactic structure in source code. Specifically, novel tree positional encodings have been proposed to incorporate inductive bias into Transformer.In this work, we propose a novel tree Transformer encoding node positions based on our new description method for tree structures.Technically, local and global soft bias shown in previous works is both introduced as positional encodings of our Transformer model.Our model finally outperforms strong baselines on code summarization and completion tasks across two languages, demonstrating our model’s effectiveness.Besides, extensive experiments and ablation study shows that combining both local and global paradigms is still helpful in improving model performance. We release our code at

RASAT: Integrating Relational Structures into Pretrained Seq2Seq Model for Text-to-SQL
Jiexing Qi | Jingyao Tang | Ziwei He | Xiangpeng Wan | Yu Cheng | Chenghu Zhou | Xinbing Wang | Quanshi Zhang | Zhouhan Lin

Relational structures such as schema linking and schema encoding have been validated as a key component to qualitatively translating natural language into SQL queries. However, introducing these structural relations comes with prices: they often result in a specialized model structure, which largely prohibits using large pretrained models in text-to-SQL. To address this problem, we propose RASAT: a Transformer seq2seq architecture augmented with relation-aware self-attention that could leverage a variety of relational structures while inheriting the pretrained parameters from the T5 model effectively. Our model can incorporate almost all types of existing relations in the literature, and in addition, we propose introducing co-reference relations for the multi-turn scenario. Experimental results on three widely used text-to-SQL datasets, covering both single-turn and multi-turn scenarios, have shown that RASAT could achieve competitive results in all three benchmarks, achieving state-of-the-art execution accuracy (75.5% EX on Spider, 52.6% IEX on SParC, and 37.4% IEX on CoSQL).

COM-MRC: A COntext-Masked Machine Reading Comprehension Framework for Aspect Sentiment Triplet Extraction
Zepeng Zhai | Hao Chen | Fangxiang Feng | Ruifan Li | Xiaojie Wang

Aspect Sentiment Triplet Extraction (ASTE) aims to extract sentiment triplets from sentences, which was recently formalized as an effective machine reading comprehension (MRC) based framework. However, when facing multiple aspect terms, the MRC-based methods could fail due to the interference from other aspect terms. In this paper, we propose a novel COntext-Masked MRC (COM-MRC) framework for ASTE. Our COM-MRC framework comprises three closely-related components: a context augmentation strategy, a discriminative model, and an inference method. Specifically, a context augmentation strategy is designed by enumerating all masked contexts for each aspect term. The discriminative model comprises four modules, i.e., aspect and opinion extraction modules, sentiment classification and aspect detection modules. In addition, a two-stage inference method first extracts all aspects and then identifies their opinions and sentiment through iteratively masking the aspects. Extensive experimental results on benchmark datasets show the effectiveness of our proposed COM-MRC framework, which outperforms state-of-the-art methods consistently.

CEM: Machine-Human Chatting Handoff via Causal-Enhance Module
Shanshan Zhong | Jinghui Qin | Zhongzhan Huang | Daifeng Li

Aiming to ensure chatbot quality by predicting chatbot failure and enabling human-agent collaboration, Machine-Human Chatting Handoff (MHCH) has attracted lots of attention from both industry and academia in recent years. However, most existing methods mainly focus on the dialogue context or assist with global satisfaction prediction based on multi-task learning, which ignore the grounded relationships among the causal variables, like the user state and labor cost. These variables are significantly associated with handoff decisions, resulting in prediction bias and cost increasement. Therefore, we propose Causal-Enhance Module (CEM) by establishing the causal graph of MHCH based on these two variables, which is a simple yet effective module and can be easy to plug into the existing MHCH methods. For the impact of users, we use the user state to correct the prediction bias according to the causal relationship of multi-task. For the labor cost, we train an auxiliary cost simulator to calculate unbiased labor cost through counterfactual learning so that a model becomes cost-aware.Extensive experiments conducted on four real-world benchmarks demonstrate the effectiveness of CEM in generally improving the performance of existing MHCH methods without any elaborated model crafting.

Nearest Neighbor Zero-Shot Inference
Weijia Shi | Julian Michael | Suchin Gururangan | Luke Zettlemoyer

Retrieval-augmented language models (LMs) use non-parametric memory to substantially outperform their non-retrieval counterparts on perplexity-based evaluations, but it is an open question whether they achieve similar gains in few- and zero-shot end-task accuracy. We extensively study one such model, the k-nearest neighbor LM (kNN-LM), showing that the gains marginally transfer. The main challenge is to achieve coverage of the verbalizer tokens that define the different end-task class labels. To address this challenge, we also introduce kNN-Prompt, a simple and effective kNN-LM with automatically expanded fuzzy verbalizers (e.g. to expand “terrible” to also include “silly” and other task-specific synonyms for sentiment classification). Across nine diverse end-tasks, using kNN-Prompt with GPT-2 large yields significant performance boosts over strong zeroshot baselines (13.4% absolute improvement over the base LM on average). We also show that other advantages of non-parametric augmentation hold for end tasks; kNN-Prompt is effective for domain adaptation with no further training, and gains increase with the size of the retrieval model.

Robots-Dont-Cry: Understanding Falsely Anthropomorphic Utterances in Dialog Systems
David Gros | Yu Li | Zhou Yu

Dialog systems are often designed or trained to output human-like responses. However, some responses may be impossible for a machine to truthfully say (e.g. “that movie made me cry”). Highly anthropomorphic responses might make users uncomfortable or implicitly deceive them into thinking they are interacting with a human. We collect human ratings on the feasibility of approximately 900 two-turn dialogs sampled from 9 diverse data sources. Ratings are for two hypothetical machine embodiments: a futuristic humanoid robot and a digital assistant. We find that for some data-sources commonly used to train dialog systems, 20-30% of utterances are not viewed as possible for a machine. Rating is marginally affected by machine embodiment. We explore qualitative and quantitative reasons for these ratings. Finally, we build classifiers and explore how modeling configuration might affect output permissibly, and discuss implications for building less falsely anthropomorphic dialog systems.

A Joint Learning Framework for Restaurant Survival Prediction and Explanation
Xin Li | Xiaojie Zhang | Peng JiaHao | Rui Mao | Mingyang Zhou | Xing Xie | Hao Liao

The bloom of the Internet and the recent breakthroughs in deep learning techniques open a new door to AI for E-commence, with a trend of evolving from using a few financial factors such as liquidity and profitability to using more advanced AI techniques to process complex and multi-modal data. In this paper, we tackle the practical problem of restaurant survival prediction. We argue that traditional methods ignore two essential respects, which are very helpful for the task: 1) modeling customer reviews and 2) jointly considering status prediction and result explanation. Thus, we propose a novel joint learning framework for explainable restaurant survival prediction based on the multi-modal data of user-restaurant interactions and users’ textual reviews. Moreover, we design a graph neural network to capture the high-order interactions and design a co-attention mechanism to capture the most informative and meaningful signal from noisy textual reviews. Our results on two datasets show a significant and consistent improvement over the SOTA techniques (average 6.8% improvement in prediction and 45.3% improvement in explanation).

Making Pretrained Language Models Good Long-tailed Learners
Chen Zhang | Lei Ren | Jingang Wang | Wei Wu | Dawei Song

Prompt-tuning has shown appealing performance in few-shot classification by virtue of its capability in effectively exploiting pre-trained knowledge. This motivates us to check the hypothesis that prompt-tuning is also a promising choice for long-tailed classification, since the tail classes are intuitively few-shot ones. To achieve this aim, we conduct empirical studies to examine the hypothesis. The results demonstrate that prompt-tuning makes pretrained language models at least good long-tailed learners. For intuitions on why prompt-tuning can achieve good performance in long-tailed classification, we carry out in-depth analyses by progressively bridging the gap between prompt-tuning and commonly used finetuning. The summary is that the classifier structure and parameterization form the key to making good long-tailed learners, in comparison with the less important input structure. Finally, we verify the applicability of our finding to few-shot classification.

UniGeo: Unifying Geometry Logical Reasoning via Reformulating Mathematical Expression
Jiaqi Chen | Tong Li | Jinghui Qin | Pan Lu | Liang Lin | Chongyu Chen | Xiaodan Liang

Geometry problem solving is a well-recognized testbed for evaluating the high-level multi-modal reasoning capability of deep models. In most existing works, two main geometry problems: calculation and proving, are usually treated as two specific tasks, hindering a deep model to unify its reasoning capability on multiple math tasks. However, in essence, these two tasks have similar problem representations and overlapped math knowledge which can improve the understanding and reasoning ability of a deep model on both two tasks. Therefore, we construct a large-scale Unified Geometry problem benchmark, UniGeo, which contains 4,998 calculation problems and 9,543 proving problems. Each proving problem is annotated with a multi-step proof with reasons and mathematical expressions. The proof can be easily reformulated as a proving sequence that shares the same formats with the annotated program sequence for calculation problems. Naturally, we also present a unified multi-task Geometric Transformer framework, Geoformer, to tackle calculation and proving problems simultaneously in the form of sequence generation, which finally shows the reasoning ability can be improved on both two tasks by unifying formulation. Furthermore, we propose a Mathematical Expression Pretraining (MEP) method that aims to predict the mathematical expressions in the problem solution, thus improving the Geoformer model. Experiments on the UniGeo demonstrate that our proposed Geoformer obtains state-of-the-art performance by outperforming task-specific model NGS with over 5.6% and 3.2% accuracies on calculation and proving problems, respectively.

Face-Sensitive Image-to-Emotional-Text Cross-modal Translation for Multimodal Aspect-based Sentiment Analysis
Hao Yang | Yanyan Zhao | Bing Qin

Aspect-level multimodal sentiment analysis, which aims to identify the sentiment of the target aspect from multimodal data, recently has attracted extensive attention in the community of multimedia and natural language processing. Despite the recent success in textual aspect-based sentiment analysis, existing models mainly focused on utilizing the object-level semantic information in the image but ignore explicitly using the visual emotional cues, especially the facial emotions. How to distill visual emotional cues and align them with the textual content remains a key challenge to solve the problem. In this work, we introduce a face-sensitive image-to-emotional-text translation (FITE) method, which focuses on capturing visual sentiment cues through facial expressions and selectively matching and fusing with the target aspect in textual modality. To the best of our knowledge, we are the first that explicitly utilize the emotional information from images in the multimodal aspect-based sentiment analysis task. Experiment results show that our method achieves state-of-the-art results on the Twitter-2015 and Twitter-2017 datasets. The improvement demonstrates the superiority of our model in capturing aspect-level sentiment in multimodal data with facial expressions.

FineD-Eval: Fine-grained Automatic Dialogue-Level Evaluation
Chen Zhang | Luis Fernando D’Haro | Qiquan Zhang | Thomas Friedrichs | Haizhou Li

Recent model-based reference-free metrics for open-domain dialogue evaluation exhibit promising correlations with human judgment. However, they either perform turn-level evaluation or look at a single dialogue quality dimension. One would expect a good evaluation metric to assess multiple quality dimensions at the dialogue level. To this end, we are motivated to propose a multi-dimensional dialogue-level metric, which consists of three sub-metrics with each targeting a specific dimension. The sub-metrics are trained with novel self-supervised objectives and exhibit strong correlations with human judgment for their respective dimensions. Moreover, we explore two approaches to combine the sub-metrics: metric ensemble and multitask learning. Both approaches yield a holistic metric that significantly outperforms individual sub-metrics. Compared to the existing state-of-the-art metric, the combined metrics achieve around 16% relative improvement on average across three high-quality dialogue-level evaluation benchmarks.

Sentence Representation Learning with Generative Objective rather than Contrastive Objective
Bohong Wu | Hai Zhao

Though offering amazing contextualized token-level representations, current pre-trained language models take less attention on accurately acquiring sentence-level representation during their self-supervised pre-training. However, contrastive objectives which dominate the current sentence representation learning bring little linguistic interpretability and no performance guarantee on downstream semantic tasks. We instead propose a novel generative self-supervised learning objective based on phrase reconstruction. To overcome the drawbacks of previous generative methods, we carefully model intra-sentence structure by breaking down one sentence into pieces of important phrases. Empirical studies show that our generative learning achieves powerful enough performance improvement and outperforms the current state-of-the-art contrastive methods not only on the STS benchmarks, but also on downstream semantic retrieval and reranking tasks. Our code is available at

RLPrompt: Optimizing Discrete Text Prompts with Reinforcement Learning
Mingkai Deng | Jianyu Wang | Cheng-Ping Hsieh | Yihan Wang | Han Guo | Tianmin Shu | Meng Song | Eric Xing | Zhiting Hu

Prompting has shown impressive success in enabling large pre-trained language models (LMs) to perform diverse NLP tasks, especially with only few downstream data. Automatically finding the optimal prompt for each task, however, is challenging. Most existing work resorts to tuning *soft* prompts (e.g., embeddings) which fall short of interpretability, reusability across LMs, and applicability when gradients are not accessible. *Discrete* prompts, on the other hand, are difficult to optimize, and are often created by “enumeration (e.g., paraphrasing)-then-selection” heuristics that do not explore the prompt space systematically. This paper proposes RLPrompt, an efficient discrete prompt optimization approach with reinforcement learning (RL). RLPrompt formulates a parameter-efficient policy network that generates the optimized discrete prompt after training with reward. To harness the complex and stochastic reward signals from the large LM environment, we incorporate effective reward stabilization that substantially enhances training efficiency. RLPrompt is flexibly applicable to different types of LMs, such as masked (e.g., BERT) and left-to-right models (e.g., GPTs), for both classification and generation tasks. Experiments on few-shot classification and unsupervised text style transfer show superior performance over a wide range of existing fine-tuning or prompting methods. Interestingly, the resulting optimized prompts are often ungrammatical gibberish text; and surprisingly, those gibberish prompts are transferrable between different LMs to retain significant performance, indicating that LM prompting may not follow human language patterns.

DisCup: Discriminator Cooperative Unlikelihood Prompt-tuning for Controllable Text Generation
Hanqing Zhang | Dawei Song

Prompt learning with immensely large Casual Language Models (CLMs) has been shown promising for attribute-controllable text generation (CTG). However, vanilla prompt tuning tends to imitate training corpus characteristics beyond the control attributes, resulting in a poor generalization ability. Moreover, it is less able to capture the relationship between different attributes, further limiting the control performance. In this paper, we propose a new CTG approach, namely DisCup, which incorporates the attribute knowledge of discriminator to optimize the control-prompts, steering a frozen CLM to produce attribute-specific texts. Specifically, the frozen CLM model, capable of producing multitudinous texts, is first used to generate the next-token candidates based on the context, so as to ensure the diversity of tokens to be predicted. Then, we leverage an attribute-discriminator to select desired/undesired tokens from those candidates, providing the inter-attribute knowledge. Finally, we bridge the above two traits by an unlikelihood objective for prompt-tuning. Extensive experimental results show that DisCup can achieve a new state-of-the-art control performance while maintaining an efficient and high-quality text generation, only relying on around 10 virtual tokens.

CPL: Counterfactual Prompt Learning for Vision and Language Models
Xuehai He | Diji Yang | Weixi Feng | Tsu-Jui Fu | Arjun Akula | Varun Jampani | Pradyumna Narayana | Sugato Basu | William Yang Wang | Xin Wang

Prompt tuning is a new few-shot transfer learning technique that only tunes the learnable prompt for pre-trained vision and language models such as CLIP. However, existing prompt tuning methods tend to learn spurious or entangled representations, which leads to poor generalization to unseen concepts.Towards non-spurious and efficient prompt learning from limited examples, this paper presents a novel Counterfactual Prompt Learning (CPL) method for vision and language models, which simultaneously employs counterfactual generation and contrastive learning in a joint optimization framework.Particularly, CPL constructs counterfactual by identifying minimal non-spurious feature change between semantically-similar positive and negative samples that causes concept change, and learns more generalizable prompt representation from both factual and counterfactual examples via contrastive learning. Extensive experiments demonstrate that CPL can obtain superior few-shot performance on different vision and language tasks than previous prompt tuning methods on CLIP. On image classification, we achieve 3.55% average relative improvement on unseen classes across seven datasets; on image-text retrieval and visual question answering, we gain up to 4.09% and 25.08% relative improvements across three few-shot scenarios on unseen test sets respectively.

Red Teaming Language Models with Language Models
Ethan Perez | Saffron Huang | Francis Song | Trevor Cai | Roman Ring | John Aslanides | Amelia Glaese | Nat McAleese | Geoffrey Irving

Language Models (LMs) often cannot be deployed because of their potential to harm users in hard-to-predict ways. Prior work identifies harmful behaviors before deployment by using human annotators to hand-write test cases. However, human annotation is expensive, limiting the number and diversity of test cases. In this work, we automatically find cases where a target LM behaves in a harmful way, by generating test cases (“red teaming”) using another LM. We evaluate the target LM’s replies to generated test questions using a classifier trained to detect offensive content, uncovering tens of thousands of offensive replies in a 280B parameter LM chatbot. We explore several methods, from zero-shot generation to reinforcement learning, for generating test cases with varying levels of diversity and difficulty. Furthermore, we use prompt engineering to control LM-generated test cases to uncover a variety of other harms, automatically finding groups of people that the chatbot discusses in offensive ways, personal and hospital phone numbers generated as the chatbot’s own contact info, leakage of private training data in generated text, and harms that occur over the course of a conversation. Overall, LM-based red teaming is one promising tool (among many needed) for finding and fixing diverse, undesirable LM behaviors before impacting users.

CapOnImage: Context-driven Dense-Captioning on Image
Yiqi Gao | Xinglin Hou | Yuanmeng Zhang | Tiezheng Ge | Yuning Jiang | Peng Wang

Existing image captioning systems are dedicated to generating narrative captions for images, which are spatially detached from theimage in presentation. However, texts can also be used as decorations on the image to highlight the key points and increase theattractiveness of images. In this work, we introduce a new taskcalled captioning on image (CapOnImage), which aims to generatedense captions at different locations of the image based on contextual information. To fully exploit the surrounding visual context togenerate the most suitable caption for each location, we propose amulti-modal pre-training model with multi-level pre-training tasksthat progressively learn the correspondence between texts and image locations from easy to difficult. Since the model may generateredundant captions for nearby locations, we further enhance thelocation embedding with neighbor locations as context. For thisnew task, we also introduce a large-scale benchmark called CapOnImage2M, which contains 2.1 million product images, each with anaverage of 4.8 spatially localized captions. Compared with other image captioning model variants, our model achieves the best resultsin both captioning accuracy and diversity aspects.

SpanProto: A Two-stage Span-based Prototypical Network for Few-shot Named Entity Recognition
Jianing Wang | Chengyu Wang | Chuanqi Tan | Minghui Qiu | Songfang Huang | Jun Huang | Ming Gao

Few-shot Named Entity Recognition (NER) aims to identify named entities with very little annotated data. Previous methods solve this problem based on token-wise classification, which ignores the information of entity boundaries, and inevitably the performance is affected by the massive non-entity tokens. To this end, we propose a seminal span-based prototypical network (SpanProto) that tackles few-shot NER via a two-stage approach, including span extraction and mention classification. In the span extraction stage, we transform the sequential tags into a global boundary matrix, enabling the model to focus on the explicit boundary information. For mention classification, we leverage prototypical learning to capture the semantic representations for each labeled span and make the model better adapt to novel-class entities. To further improve the model performance, we split out the false positives generated by the span extractor but not labeled in the current episode set, and then present a margin-based loss to separate them from each prototype region. Experiments over multiple benchmarks demonstrate that our model outperforms strong baselines by a large margin.

Discovering Differences in the Representation of People using Contextualized Semantic Axes
Li Lucy | Divya Tadimeti | David Bamman

A common paradigm for identifying semantic differences across social and temporal contexts is the use of static word embeddings and their distances. In particular, past work has compared embeddings against “semantic axes” that represent two opposing concepts. We extend this paradigm to BERT embeddings, and construct contextualized axes that mitigate the pitfall where antonyms have neighboring representations. We validate and demonstrate these axes on two people-centric datasets: occupations from Wikipedia, and multi-platform discussions in extremist, men’s communities over fourteen years. In both studies, contextualized semantic axes can characterize differences among instances of the same word type. In the latter study, we show that references to women and the contexts around them have become more detestable over time.

Generating Literal and Implied Subquestions to Fact-check Complex Claims
Jifan Chen | Aniruddh Sriram | Eunsol Choi | Greg Durrett

Verifying political claims is a challenging task, as politicians can use various tactics to subtly misrepresent the facts for their agenda. Existing automatic fact-checking systems fall short here, and their predictions like “half-true” are not very useful in isolation, since it is unclear which parts of a claim are true and which are not. In this work, we focus on decomposing a complex claim into a comprehensive set of yes-no subquestions whose answers influence the veracity of the claim. We present CLAIMDECOMP, a dataset of decompositions for over 1000 claims. Given a claim and its verification paragraph written by fact-checkers, our trained annotators write subquestions covering both explicit propositions of the original claim and its implicit facets, such as asking about additional political context that changes our view of the claim’s veracity. We study whether state-of-the-art models can generate such subquestions, showing that these models generate reasonable questions to ask, but predicting the comprehensive set of subquestions from the original claim without evidence remains challenging. We further show that these subquestions can help identify relevant evidence to fact-check the full claim and derive the veracity through their answers, suggesting that they can be useful pieces of a fact-checking pipeline.

Machine Translation Robustness to Natural Asemantic Variation
Jacob Bremerman | Xiang Ren | Jonathan May

Current Machine Translation (MT) models still struggle with more challenging input, such as noisy data and tail-end words and phrases. Several works have addressed this robustness issue by identifying specific categories of noise and variation then tuning models to perform better on them. An important yet under-studied category involves minor variations in nuance (non-typos) that preserve meaning w.r.t. the target language. We introduce and formalize this category as Natural Asemantic Variation (NAV) and investigate it in the context of MT robustness. We find that existing MT models fail when presented with NAV data, but we demonstrate strategies to improve performance on NAV by fine-tuning them with human-generated variations. We also show that NAV robustness can be transferred across languages and find that synthetic perturbations can achieve some but not all of the benefits of organic NAV data.

Natural Language to Code Translation with Execution
Freda Shi | Daniel Fried | Marjan Ghazvininejad | Luke Zettlemoyer | Sida I. Wang

Generative models of code, pretrained on large corpora of programs, have shown great success in translating natural language to code (Chen et al., 2021; Austin et al., 2021; Li et al., 2022, inter alia). While these models do not explicitly incorporate program semantics (i.e., execution results) during training, they are able to generate correct solutions for many problems. However, choosing a single correct program from a generated set for each problem remains challenging. In this work, we introduce execution result–based minimum Bayes risk decoding (MBR-EXEC) for program selection and show that it improves the few-shot performance of pretrained code models on natural-language-to-code tasks. We select output programs from a generated candidate set by marginalizing over program implementations that share the same semantics. Because exact equivalence is intractable, we execute each program on a small number of test inputs to approximate semantic equivalence. Across datasets, execution or simulated execution significantly outperforms the methods that do not involve program semantics. We find that MBR-EXEC consistently improves over all execution-unaware selection methods, suggesting it as an effective approach for natural language to code translation.

Life is a Circus and We are the Clowns: Automatically Finding Analogies between Situations and Processes
Oren Sultan | Dafna Shahaf

Analogy-making gives rise to reasoning, abstraction, flexible categorization and counterfactual inference – abilities lacking in even the best AI systems today. Much research has suggested that analogies are key to non-brittle systems that can adapt to new domains. Despite their importance, analogies received little attention in the NLP community, with most research focusing on simple word analogies. Work that tackled more complex analogies relied heavily on manually constructed, hard-to-scale input representations.In this work, we explore a more realistic, challenging setup: our input is a pair of natural language procedural texts, describing a situation or a process (e.g., how the heart works/how a pump works). Our goal is to automatically extract entities and their relations from the text and find a mapping between the different domains based on relational similarity (e.g., blood is mapped to water). We develop an interpretable, scalable algorithm and demonstrate that it identifies the correct mappings 87% of the time for procedural texts and 94% for stories from cognitive-psychology literature. We show it can extract analogies from a large dataset of procedural texts, achieving 79% precision (analogy prevalence in data: 3%). Lastly, we demonstrate that our algorithm is robust to paraphrasing the input texts

Language Contamination Helps Explains the Cross-lingual Capabilities of English Pretrained Models
Terra Blevins | Luke Zettlemoyer

English pretrained language models, which make up the backbone of many modern NLP systems, require huge amounts of unlabeled training data. These models are generally presented as being trained only on English text but have been found to transfer surprisingly well to other languages. We investigate this phenomenon and find that common English pretraining corpora actually contain significant amounts of non-English text: even when less than 1% of data is not English (well within the error rate of strong language classifiers), this leads to hundreds of millions of foreign language tokens in large-scale datasets. We then demonstrate that even these small percentages of non-English data facilitate cross-lingual transfer for models trained on them, with target language performance strongly correlated to the amount of in-language data seen during pretraining. In light of these findings, we argue that no model is truly monolingual when pretrained at scale, which should be considered when evaluating cross-lingual transfer.

Analyzing the Mono- and Cross-Lingual Pretraining Dynamics of Multilingual Language Models
Terra Blevins | Hila Gonen | Luke Zettlemoyer

The emergent cross-lingual transfer seen in multilingual pretrained models has sparked significant interest in studying their behavior. However, because these analyses have focused on fully trained multilingual models, little is known about the dynamics of the multilingual pretraining process. We investigate when these models acquire their in-language and cross-lingual abilities by probing checkpoints taken from throughout XLM-R pretraining, using a suite of linguistic tasks. Our analysis shows that the model achieves high in-language performance early on, with lower-level linguistic skills acquired before more complex ones. In contrast, the point in pretraining when the model learns to transfer cross-lingually differs across language pairs. Interestingly, we also observe that, across many languages and tasks, the final model layer exhibits significant performance degradation over time, while linguistic knowledge propagates to lower layers of the network. Taken together, these insights highlight the complexity of multilingual pretraining and the resulting varied behavior for different languages over time.

Neural Machine Translation with Contrastive Translation Memories
Xin Cheng | Shen Gao | Lemao Liu | Dongyan Zhao | Rui Yan

Retrieval-augmented Neural Machine Translation models have been successful in many translation scenarios. Different from previous works that make use of mutually similar but redundant translation memories (TMs), we propose a new retrieval-augmented NMT to model contrastively retrieved translation memories that are holistically similar to the source sentence while individually contrastive to each other providing maximal information gain in three phases. First, in TM retrieval phase, we adopt contrastive retrieval algorithm to avoid redundancy and uninformativeness of similar translation pieces. Second, in memory encoding stage, given a set of TMs we propose a novel Hierarchical Group Attention module to gather both local context of each TM and global context of the whole TM set. Finally, in training phase, a Multi-TM contrastive learning objective is introduced to learn salient feature of each TM with respect to target sentence. Experimental results show that our framework obtains substantial improvements over strong baselines in the benchmark dataset.

Distilling Causal Effect from Miscellaneous Other-Class for Continual Named Entity Recognition
Junhao Zheng | Zhanxian Liang | Haibin Chen | Qianli Ma

Continual Learning for Named Entity Recognition (CL-NER) aims to learn a growing number of entity types over time from a stream of data. However, simply learning Other-Class in the same way as new entity types amplifies the catastrophic forgetting and leads to a substantial performance drop. The main cause behind this is that Other-Class samples usually contain old entity types, and the old knowledge in these Other-Class samples is not preserved properly. Thanks to the causal inference, we identify that the forgetting is caused by the missing causal effect from the old data.To this end, we propose a unified causal framework to retrieve the causality from both new entity types and Other-Class.Furthermore, we apply curriculum learning to mitigate the impact of label noise and introduce a self-adaptive weight for balancing the causal effects between new entity types and Other-Class. Experimental results on three benchmark datasets show that our method outperforms the state-of-the-art method by a large margin. Moreover, our method can be combined with the existing state-of-the-art methods to improve the performance in CL-NER.

Exploring the Secrets Behind the Learning Difficulty of Meaning Representations for Semantic Parsing
Zhenwen Li | Jiaqi Guo | Qian Liu | Jian-Guang Lou | Tao Xie

Previous research has shown that the design of Meaning Representation (MR) greatly influences the final model performance of a neural semantic parser. Therefore, designing a good MR is a long-term goal for semantic parsing. However, it is still an art as there is no quantitative indicator that can tell us which MR among a set of candidates may have the best final model performance. In practice, in order toselect an MR design, researchers often have to go through the whole training-testing process for all design candidates, and the process often costs a lot. In this paper, we propose a data-aware metric called ISS (denoting incremental structural stability) of MRs, and demonstrate that ISS is highly correlated with the final performance. The finding shows that ISS can be used as an indicator for MR design to avoid the costly training-testing process.

That’s the Wrong Lung! Evaluating and Improving the Interpretability of Unsupervised Multimodal Encoders for Medical Data
Jered McInerney | Geoffrey Young | Jan-Willem van de Meent | Byron Wallace

Pretraining multimodal models on Electronic Health Records (EHRs) provides a means of learning representations that can transfer to downstream tasks with minimal supervision. Recent multimodal models induce soft local alignments between image regions and sentences. This is of particular interest in the medical domain, where alignments might highlight regions in an image relevant to specific phenomena described in free-text. While past work has suggested that attention “heatmaps” can be interpreted in this manner, there has been little evaluation of such alignments. We compare alignments from a state-of-the-art multimodal (image and text) model for EHR with human annotations that link image regions to sentences. Our main finding is that the text has an often weak or unintuitive influence on attention; alignments do not consistently reflect basic anatomical information. Moreover, synthetic modifications — such as substituting “left” for “right” — do not substantially influence highlights. Simple techniques such as allowing the model to opt out of attending to the image and few-shot finetuning show promise in terms of their ability to improve alignments with very little or no supervision. We make our code and checkpoints open-source.

Unsupervised Tokenization Learning
Anton Kolonin | Vignav Ramesh

In the presented study, we discover that the so-called “transition freedom” metric appears superior for unsupervised tokenization purposes in comparison to statistical metrics such as mutual information and conditional probability, providing F-measure scores in range from 0.71 to 1.0 across explored multilingual corpora. We find that different languages require different offshoots of that metric (such as derivative, variance, and “peak values”) for successful tokenization. Larger training corpora do not necessarily result in better tokenization quality, while compressing the models by eliminating statistically weak evidence tends to improve performance. The proposed unsupervised tokenization technique provides quality better than or comparable to lexicon-based ones, depending on the language.

A Template-based Method for Constrained Neural Machine Translation
Shuo Wang | Peng Li | Zhixing Tan | Zhaopeng Tu | Maosong Sun | Yang Liu

Machine translation systems are expected to cope with various types of constraints in many practical scenarios. While neural machine translation (NMT) has achieved strong performance in unconstrained cases, it is non-trivial to impose pre-specified constraints into the translation process of NMT models. Although many approaches have been proposed to address this issue, most existing methods can not satisfy the following three desiderata at the same time: (1) high translation quality, (2) high match accuracy, and (3) low latency. In this work, we propose a template-based method that can yield results with high translation quality and match accuracy and the inference speed of our method is comparable with unconstrained NMT models. Our basic idea is to rearrange the generation of constrained and unconstrained tokens through a template. Our method does not require any changes in the model architecture and the decoding algorithm. Experimental results show that the proposed template-based approach can outperform several representative baselines in both lexically and structurally constrained translation tasks.

PATS: Sensitivity-aware Noisy Learning for Pretrained Language Models
Yupeng Zhang | Hongzhi Zhang | Sirui Wang | Wei Wu | Zhoujun Li

A wide range of NLP tasks benefit from the fine-tuning of pretrained language models (PLMs). However, a number of redundant parameters which contribute less to the downstream task are observed in a directly fine-tuned model. We consider the gap between pretraining and downstream tasks hinders the training of these redundant parameters, and results in a suboptimal performance of the overall model. In this paper, we present PATS (Perturbation According To Sensitivity), a noisy training mechanism which considers each parameter’s importance in the downstream task to help fine-tune PLMs. The main idea of PATS is to add bigger noise to parameters with lower sensitivity and vice versa, in order to activate more parameters’ contributions to downstream tasks without affecting the sensitive ones much. Extensive experiments conducted on different tasks of the GLUE benchmark show PATS can consistently empower the fine-tuning of different sizes of PLMs, and the parameters in the well-performing models always have more concentrated distributions of sensitivities, which experimentally proves the effectiveness of our method.

Towards Reinterpreting Neural Topic Models via Composite Activations
Jia Peng Lim | Hady Lauw

Most Neural Topic Models (NTM) use a variational auto-encoder framework producing K topics limited to the size of the encoder’s output. These topics are interpreted through the selection of the top activated words via the weights or reconstructed vector of the decoder that are directly connected to each neuron. In this paper, we present a model-free two-stage process to reinterpret NTM and derive further insights on the state of the trained model. Firstly, building on the original information from a trained NTM, we generate a pool of potential candidate “composite topics” by exploiting possible co-occurrences within the original set of topics, which decouples the strict interpretation of topics from the original NTM. This is followed by a combinatorial formulation to select a final set of composite topics, which we evaluate for coherence and diversity on a large external corpus. Lastly, we employ a user study to derive further insights on the reinterpretation process.

Few-shot Query-Focused Summarization with Prefix-Merging
Ruifeng Yuan | Zili Wang | Ziqiang Cao | Wenjie Li

Query-focused summarization has been considered as an important extension for text summarization. It aims to generate a concise highlight for a given query. Different from text summarization, query-focused summarization has long been plagued by the problem of lacking high-quality large-scale datasets. In this paper, we investigate the idea that whether we can integrate and transfer the knowledge of text summarization and question answering to assist the few-shot learning in query-focused summarization. Here, we propose prefix-merging, a prefix-based pretraining strategy for few-shot learning in query-focused summarization. Drawn inspiration from prefix-tuning, we are allowed to integrate the task knowledge from text summarization and question answering into a properly designed prefix and apply the merged prefix to query-focused summarization. With only a small amount of trainable parameters, prefix-merging outperforms fine-tuning on query-focused summarization. We further discuss the influence of different prefix designs and propose a visualized explanation for how prefix-merging works.

Cross-Align: Modeling Deep Cross-lingual Interactions for Word Alignment
Siyu Lai | Zhen Yang | Fandong Meng | Yufeng Chen | Jinan Xu | Jie Zhou

Word alignment which aims to extract lexicon translation equivalents between source and target sentences, serves as a fundamental tool for natural language processing. Recent studies in this area have yielded substantial improvements by generating alignments from contextualized embeddings of the pre-trained multilingual language models. However, we find that the existing approaches capture few interactions between the input sentence pairs, which degrades the word alignment quality severely, especially for the ambiguous words in the monolingual context. To remedy this problem, we propose Cross-Align to model deep interactions between the input sentence pairs, in which the source and target sentences are encoded separately with the shared self-attention modules in the shallow layers, while cross-lingual interactions are explicitly constructed by the cross-attention modules in the upper layers. Besides, to train our model effectively, we propose a two-stage training framework, where the model is trained with a simple Translation Language Modeling (TLM) objective in the first stage and then finetuned with a self-supervised alignment objective in the second stage. Experiments show that the proposed Cross-Align achieves the state-of-the-art (SOTA) performance on four out of five language pairs.

BERTScore is Unfair: On Social Bias in Language Model-Based Metrics for Text Generation
Tianxiang Sun | Junliang He | Xipeng Qiu | Xuanjing Huang

Automatic evaluation metrics are crucial to the development of generative systems. In recent years, pre-trained language model (PLM) based metrics, such as BERTScore, have been commonly adopted in various generation tasks. However, it has been demonstrated that PLMs encode a range of stereotypical societal biases, leading to a concern about the fairness of PLMs as metrics. To that end, this work presents the first systematic study on the social bias in PLM-based metrics. We demonstrate that popular PLM-based metrics exhibit significantly higher social bias than traditional metrics on 6 sensitive attributes, namely race, gender, religion, physical appearance, age, and socioeconomic status. In-depth analysis suggests that choosing paradigms (matching, regression, or generation) of the metric has a greater impact on fairness than choosing PLMs. In addition, we develop debiasing adapters that are injected into PLM layers, mitigating bias in PLM-based metrics while retaining high performance for evaluating text generation.

HPT: Hierarchy-aware Prompt Tuning for Hierarchical Text Classification
Zihan Wang | Peiyi Wang | Tianyu Liu | Binghuai Lin | Yunbo Cao | Zhifang Sui | Houfeng Wang

Hierarchical text classification (HTC) is a challenging subtask of multi-label classification due to its complex label hierarchy.Recently, the pretrained language models (PLM)have been widely adopted in HTC through a fine-tuning paradigm. However, in this paradigm, there exists a huge gap between the classification tasks with sophisticated label hierarchy and the masked language model (MLM) pretraining tasks of PLMs and thus the potential of PLMs cannot be fully tapped.To bridge the gap, in this paper, we propose HPT, a Hierarchy-aware Prompt Tuning method to handle HTC from a multi-label MLM perspective.Specifically, we construct a dynamic virtual template and label words that take the form of soft prompts to fuse the label hierarchy knowledge and introduce a zero-bounded multi-label cross-entropy loss to harmonize the objectives of HTC and MLM.Extensive experiments show HPT achieves state-of-the-art performances on 3 popular HTC datasets and is adept at handling the imbalance and low resource situations. Our code is available at

Not to Overfit or Underfit the Source Domains? An Empirical Study of Domain Generalization in Question Answering
Md Arafat Sultan | Avi Sil | Radu Florian

Machine learning models are prone to overfitting their training (source) domains, which is commonly believed to be the reason why they falter in novel target domains. Here we examine the contrasting view that multi-source domain generalization (DG) is first and foremost a problem of mitigating source domain underfitting: models not adequately learning the signal already present in their multi-domain training data. Experiments on a reading comprehension DG benchmark show that as a model learns its source domains better—using familiar methods such as knowledge distillation (KD) from a bigger model—its zero-shot out-of-domain utility improves at an even faster pace. Improved source domain learning also demonstrates superior out-of-domain generalization over three popular existing DG approaches that aim to limit overfitting. Our implementation of KD-based domain generalization is available via PrimeQA at:

Neural Theory-of-Mind? On the Limits of Social Intelligence in Large LMs
Maarten Sap | Ronan Le Bras | Daniel Fried | Yejin Choi

Social intelligence and Theory of Mind (TOM), i.e., the ability to reason about the different mental states, intents, and reactions of all people involved, allows humans to effectively navigate and understand everyday social interactions. As NLP systems are used in increasingly complex social situations, their ability to grasp social dynamics becomes crucial.In this work, we examine the open question of social intelligence and Theory of Mind in modern NLP systems from an empirical and theorybased perspective. We show that one of today’s largest language models (GPT-3; Brown et al., 2020) lacks this kind of social intelligence out-of-the box, using two tasks: SocialIQa (Sap et al., 2019), which measure models’ ability to understand intents and reactions of participants of social interactions, and ToMi (Le, Boureau, and Nickel, 2019), which measures whether models can infer mental states and realities of participants of situations.Our results show that models struggle substantially at these Theory of Mind tasks, with well-below-human accuracies of 55% and 60% on SocialIQa and ToMi, respectively. To conclude, we draw on theories from pragmatics to contextualize this shortcoming of large language models, by examining the limitations stemming from their data, neural architecture, and training paradigms. Challenging the prevalent narrative that only scale is needed, we posit that person-centric NLP approaches might be more effective towards neural Theory of Mind.

Improving Passage Retrieval with Zero-Shot Question Generation
Devendra Sachan | Mike Lewis | Mandar Joshi | Armen Aghajanyan | Wen-tau Yih | Joelle Pineau | Luke Zettlemoyer

We propose a simple and effective re-ranking method for improving passage retrieval in open question answering. The re-ranker re-scores retrieved passages with a zero-shot question generation model, which uses a pre-trained language model to compute the probability of the input question conditioned on a retrieved passage. This approach can be applied on top of any retrieval method (e.g. neural or keyword-based), does not require any domain- or task-specific training (and therefore is expected to generalize better to data distribution shifts), and provides rich cross-attention between query and passage (i.e. it must explain every token in the question). When evaluated on a number of open-domain retrieval datasets, our re-ranker improves strong unsupervised retrieval models by 6%-18% absolute and strong supervised models by up to 12% in terms of top-20 passage retrieval accuracy. We also obtain new state-of-the-art results on full open-domain question answering by simply adding the new re-ranker to existing models with no further changes.

Summarizing Community-based Question-Answer Pairs
Ting-Yao Hsu | Yoshi Suhara | Xiaolan Wang

Community-based Question Answering (CQA), which allows users to acquire their desired information, has increasingly become an essential component of online services in various domains such as E-commerce, travel, and dining. However, an overwhelming number of CQA pairs makes it difficult for users without particular intent to find useful information spread over CQA pairs. To help users quickly digest the key information, we propose the novel CQA summarization task that aims to create a concise summary from CQA pairs. To this end, we first design a multi-stage data annotation process and create a benchmark dataset, COQASUM, based on the Amazon QA corpus. We then compare a collection of extractive and abstractive summarization methods and establish a strong baseline approach DedupLED for the CQA summarization task. Our experiment further confirms two key challenges, sentence-type transfer and deduplication removal, towards the CQA summarization task. Our data and code are publicly available.

Logical Reasoning with Span-Level Predictions for Interpretable and Robust NLI Models
Joe Stacey | Pasquale Minervini | Haim Dubossarsky | Marek Rei

Current Natural Language Inference (NLI) models achieve impressive results, sometimes outperforming humans when evaluating on in-distribution test sets. However, as these models are known to learn from annotation artefacts and dataset biases, it is unclear to what extent the models are learning the task of NLI instead of learning from shallow heuristics in their training data.We address this issue by introducing a logical reasoning framework for NLI, creating highly transparent model decisions that are based on logical rules. Unlike prior work, we show that improved interpretability can be achieved without decreasing the predictive accuracy. We almost fully retain performance on SNLI, while also identifying the exact hypothesis spans that are responsible for each model prediction.Using the e-SNLI human explanations, we verify that our model makes sensible decisions at a span level, despite not using any span labels during training. We can further improve model performance and the span-level decisions by using the e-SNLI explanations during training. Finally, our model is more robust in a reduced data setting. When training with only 1,000 examples, out-of-distribution performance improves on the MNLI matched and mismatched validation sets by 13% and 16% relative to the baseline. Training with fewer observations yields further improvements, both in-distribution and out-of-distribution.

How to disagree well: Investigating the dispute tactics used on Wikipedia
Christine De Kock | Andreas Vlachos

Disagreements are frequently studied from the perspective of either detecting toxicity or analysing argument structure. We propose a framework of dispute tactics which unifies these two perspectives, as well as other dialogue acts which play a role in resolving disputes, such as asking questions and providing clarification. This framework includes a preferential ordering among rebuttal-type tactics, ranging from ad hominem attacks to refuting the central argument. Using this framework, we annotate 213 disagreements (3,865 utterances) from Wikipedia Talk pages. This allows us to investigate research questions around the tactics used in disagreements; for instance, we provide empirical validation of the approach to disagreement recommended by Wikipedia. We develop models for multilabel prediction of dispute tactics in an utterance, achieving the best performance with a transformer-based label powerset model. Adding an auxiliary task to incorporate the ordering of rebuttal tactics further yields a statistically significant increase. Finally, we show that these annotations can be used to provide useful additional signals to improve performance on the task of predicting escalation.

Chapter Ordering in Novels
Allen Kim | Steve Skiena

Understanding narrative flow and text coherence in long-form documents (novels) remains an open problem in NLP.To gain insight, we explore the task of chapter ordering, reconstructing the original order of chapters in novel given a random permutation of the text. This can be seen as extending the well-known sentence ordering task to vastly larger documents: our task deals with over 9,000 novels with an average of twenty chapters each, versus standard sentence ordering datasets averaging only 5-8 sentences. We formulate the task of reconstructing order as a constraint solving problem, using minimum feedback arc set and traveling salesman problem optimization criteria, where the weights of the graph are generated based on models for character occurrences and chapter boundary detection, using relational chapter scores derived from RoBERTa. Our best methods yield a Spearman correlation of 0.59 on this novel and challenging task, substantially above baseline.

Open-ended Knowledge Tracing for Computer Science Education
Naiming Liu | Zichao Wang | Richard Baraniuk | Andrew Lan

In educational applications, knowledge tracing refers to the problem of estimating students’ time-varying concept/skill mastery level from their past responses to questions and predicting their future performance.One key limitation of most existing knowledge tracing methods is that they treat student responses to questions as binary-valued, i.e., whether they are correct or incorrect. Response correctness analysis/prediction is straightforward, but it ignores important information regarding mastery, especially for open-ended questions.In contrast, exact student responses can provide much more information.In this paper, we conduct the first exploration int open-ended knowledge tracing (OKT) by studying the new task of predicting students’ exact open-ended responses to questions.Our work is grounded in the domain of computer science education with programming questions. We develop an initial solution to the OKT problem, a student knowledge-guided code generation approach, that combines program synthesis methods using language models with student knowledge tracing methods. We also conduct a series of quantitative and qualitative experiments on a real-world student code dataset to validate and demonstrate the promise of OKT.

Logical Neural Networks for Knowledge Base Completion with Embeddings & Rules
Prithviraj Sen | Breno William Carvalho | Ibrahim Abdelaziz | Pavan Kapanipathi | Salim Roukos | Alexander Gray

Knowledge base completion (KBC) has benefitted greatly by learning explainable rules in an human-interpretable dialect such as first-order logic. Rule-based KBC has so far, mainly focussed on learning one of two types of rules: conjunction-of-disjunctions and disjunction-of-conjunctions. We qualitatively show, via examples, that one of these has an advantage over the other when it comes to achieving high quality KBC. To the best of our knowledge, we are the first to propose learning both kinds of rules within a common framework. To this end, we propose to utilize logical neural networks (LNN), a powerful neuro-symbolic AI framework that can express both kinds of rules and learn these end-to-end using gradient-based optimization. Our in-depth experiments show that our LNN-based approach to learning rules for KBC leads to roughly 10% relative improvements, if not more, over SotA rule-based KBC methods. Moreover, by showing how to combine our proposed methods with knowledge graph embeddings we further achieve an additional 7.5% relative improvement.

MedCLIP: Contrastive Learning from Unpaired Medical Images and Text
Zifeng Wang | Zhenbang Wu | Dinesh Agarwal | Jimeng Sun

Existing vision-text contrastive learning like CLIP aims to match the paired image and caption embeddings while pushing others apart, which improves representation transferability and supports zero-shot prediction. However, medical image-text datasets are orders of magnitude below the general images and captions from the internet. Moreover, previous methods encounter many false negatives, i.e., images and reports from separate patients probably carry the same semantics but are wrongly treated as negatives. In this paper, we decouple images and texts for multimodal contrastive learning, thus scaling the usable training data in a combinatorial magnitude with low cost. We also propose to replace the InfoNCE loss with semantic matching loss based on medical knowledge to eliminate false negatives in contrastive learning. We prove that MedCLIP is a simple yet effective framework: it outperforms state-of-the-art methods on zero-shot prediction, supervised classification, and image-text retrieval. Surprisingly, we observe that with only 20K pre-training data, MedCLIP wins over the state-of-the-art method (using 200K data). The code is available at

GA-SAM: Gradient-Strength based Adaptive Sharpness-Aware Minimization for Improved Generalization
Zhiyuan Zhang | Ruixuan Luo | Qi Su | Xu Sun

Recently, Sharpness-Aware Minimization (SAM) algorithm has shown state-of-the-art generalization abilities in vision tasks. It demonstrates that flat minima tend to imply better generalization abilities. However, it has some difficulty implying SAM to some natural language tasks, especially to models with drastic gradient changes, such as RNNs. In this work, we analyze the relation between the flatness of the local minimum and its generalization ability from a novel and straightforward theoretical perspective. We propose that the shift of the training and test distributions can be equivalently seen as a virtual parameter corruption or perturbation, which can explain why flat minima that are robust against parameter corruptions or perturbations have better generalization performances. On its basis, we propose a Gradient-Strength based Adaptive Sharpness-Aware Minimization (GA-SAM) algorithm to help to learn algorithms find flat minima that generalize better. Results in various language benchmarks validate the effectiveness of the proposed GA-SAM algorithm on natural language tasks.

Sparse Teachers Can Be Dense with Knowledge
Yi Yang | Chen Zhang | Dawei Song

Recent advances in distilling pretrained language models have discovered that, besides the expressiveness of knowledge, the student-friendliness should be taken into consideration to realize a truly knowledgeable teacher. Based on a pilot study, we find that over-parameterized teachers can produce expressive yet student-unfriendly knowledge and are thus limited in overall knowledgeableness. To remove the parameters that result in student-unfriendliness, we propose a sparse teacher trick under the guidance of an overall knowledgeable score for each teacher parameter. The knowledgeable score is essentially an interpolation of the expressiveness and student-friendliness scores. The aim is to ensure that the expressive parameters are retained while the student-unfriendly ones are removed. Extensive experiments on the GLUE benchmark show that the proposed sparse teachers can be dense with knowledge and lead to students with compelling performance in comparison with a series of competitive baselines.

BBTv2: Towards a Gradient-Free Future with Large Language Models
Tianxiang Sun | Zhengfu He | Hong Qian | Yunhua Zhou | Xuanjing Huang | Xipeng Qiu

Most downstream adaptation methods tune all or part of the parameters of pre-trained models (PTMs) through gradient descent, where the tuning cost increases linearly with the growth of the model size.By contrast, gradient-free methods only require the forward computation of the PTM to tune the prompt, retaining the benefits of efficient tuning and deployment.Though, past work on gradient-free tuning often introduces gradient descent to seek a good initialization of prompt and lacks versatility across tasks and PTMs.In this paper, we present BBTv2, an improved version of Black-Box Tuning, to drive PTMs for few-shot learning.We prepend continuous prompts to every layer of the PTM and propose a divide-and-conquer gradient-free algorithm to optimize the prompts at different layers alternately.Extensive experiments across various tasks and PTMs show that BBTv2 can achieve comparable performance to full model tuning and state-of-the-art parameter-efficient methods (e.g., Adapter, LoRA, BitFit, etc.) under few-shot settings while maintaining much fewer tunable parameters.

Passage-Mask: A Learnable Regularization Strategy for Retriever-Reader Models
Shujian Zhang | Chengyue Gong | Xingchao Liu

Retriever-reader models achieve competitive performance across many different NLP tasks such as open question answering and dialogue conversations. In this work, we notice these models easily overfit the top-rank retrieval passages and standard training fails to reason over the entire retrieval passages. We introduce a learnable passage mask mechanism which desensitizes the impact from the top-rank retrieval passages and prevents the model from overfitting. Controlling the gradient variance with fewer mask candidates and selecting the mask candidates with one-shot bi-level optimization, our learnable regularization strategy enforces the answer generation to focus on the entire retrieval passages. Experiments on different tasks across open question answering, dialogue conversation, and fact verification show that our method consistently outperforms its baselines. Extensive experiments and ablation studies demonstrate that our method can be general, effective, and beneficial for many NLP tasks.

Mixed-effects transformers for hierarchical adaptation
Julia White | Noah Goodman | Robert Hawkins

Language differs dramatically from context to context. To some degree, large language models like GPT-3 account for such variation by conditioning on strings of initial input text, or prompts. However, prompting can be ineffective when contexts are sparse, out-of-sample, or extra-textual. In this paper, we introduce the mixed-effects transformer (MET), a novel approach for learning hierarchically-structured prefixes— lightweight modules prepended to an input sequence— to account for structured variation in language use. Specifically, we show how the popular class of mixed-effects regression models may be extended to transformer-based architectures using a regularized prefix-tuning procedure with dropout. We evaluate this approach on several domain-adaptation benchmarks, finding that it learns contextual variation from minimal data while generalizing well to unseen contexts.

On Measuring the Intrinsic Few-Shot Hardness of Datasets
Xinran Zhao | Shikhar Murty | Christopher Manning

While advances in pre-training have led to dramatic improvements in few-shot learning of NLP tasks, there is limited understanding of what drives successful few-shot adaptation in datasets. In particular, given a new dataset and a pre-trained model, what properties of the dataset make it few-shot learnable, and are these properties independent of the specific adaptation techniques used? We consider an extensive set of recent few-shot learning methods and show that their performance across a large number of datasets is highly correlated, showing that few-shot hardness may be intrinsic to datasets, for a given pre-trained model. To estimate intrinsic few-shot hardness, we then propose a simple and lightweight metric called Spread that captures the intuition that few-shot learning is made possible by exploiting feature-space invariances between training and test samples. Our metric better accounts for few-shot hardness compared to existing notions of hardness and is ~8-100x faster to compute.

Group is better than individual: Exploiting Label Topologies and Label Relations for Joint Multiple Intent Detection and Slot Filling
Bowen Xing | Ivor Tsang

Recent joint multiple intent detection and slot filling models employ label embeddings to achieve the semantics-label interactions.However, they treat all labels and label embeddings as uncorrelated individuals, ignoring the dependencies among them. Besides, they conduct the decoding for the two tasks independently, without leveraging the correlations between them.Therefore, in this paper, we first construct a Heterogeneous Label Graph (HLG) containing two kinds of topologies: (1) statistical dependencies based on labels’ co-occurrence patterns and hierarchies in slot labels; (2) rich relations among the label nodes.Then we propose a novel model termed ReLa-Net.It can capture beneficial correlations among the labels from HLG.The label correlations are leveraged to enhance semantic-label interactions. Moreover, we also propose the label-aware inter-dependent decoding mechanism to further exploit the label correlations for decoding. Experiment results show that our ReLa-Net significantly outperforms previous models.Remarkably, ReLa-Net surpasses the previous best model by over 20% in terms of overall accuracy on MixATIS dataset.

An Empirical Study on Finding Spans
Weiwei Gu | Boyuan Zheng | Yunmo Chen | Tongfei Chen | Benjamin Van Durme

We present an empirical study on methods for span finding, the selection of consecutive tokens in text for some downstream tasks. We focus on approaches that can be employed in training end-to-end information extraction systems, and find there is no definitive solution without considering task properties, and provide our observations to help with future design choices: 1) a tagging approach often yields higher precision while span enumeration and boundary prediction provide higher recall; 2) span type information can benefit a boundary prediction approach; 3) additional contextualization does not help span finding in most cases.

MGDoc: Pre-training with Multi-granular Hierarchy for Document Image Understanding
Zilong Wang | Jiuxiang Gu | Chris Tensmeyer | Nikolaos Barmpalios | Ani Nenkova | Tong Sun | Jingbo Shang | Vlad Morariu

Document images are a ubiquitous source of data where the text is organized in a complex hierarchical structure ranging from fine granularity (e.g., words), medium granularity (e.g., regions such as paragraphs or figures), to coarse granularity (e.g., the whole page). The spatial hierarchical relationships between content at different levels of granularity are crucial for document image understanding tasks. Existing methods learn features from either word-level or region-level but fail to consider both simultaneously. Word-level models are restricted by the fact that they originate from pure-text language models, which only encode the word-level context. In contrast, region-level models attempt to encode regions corresponding to paragraphs or text blocks into a single embedding, but they perform worse with additional word-level features. To deal with these issues, we propose MGDoc, a new multi-modal multi-granular pre-training framework that encodes page-level, region-level, and word-level information at the same time. MGDoc uses a unified text-visual encoder to obtain multi-modal features across different granularities, which makes it possible to project the multi-granular features into the same hyperspace. To model the region-word correlation, we design a cross-granular attention mechanism and specific pre-training tasks for our model to reinforce the model of learning the hierarchy between regions and words. Experiments demonstrate that our proposed model can learn better features that perform well across granularities and lead to improvements in downstream tasks.

Understanding Jargon: Combining Extraction and Generation for Definition Modeling
Jie Huang | Hanyin Shao | Kevin Chen-Chuan Chang | Jinjun Xiong | Wen-mei Hwu

Can machines know what twin prime is? From the composition of this phrase, machines may guess twin prime is a certain kind of prime, but it is still difficult to deduce exactly what twin stands for without additional knowledge. Here, twin prime is a jargon - a specialized term used by experts in a particular field. Explaining jargon is challenging since it usually requires domain knowledge to understand. Recently, there is an increasing interest in extracting and generating definitions of words automatically. However, existing approaches, either extraction or generation, perform poorly on jargon. In this paper, we propose to combine extraction and generation for jargon definition modeling: first extract self- and correlative definitional information of target jargon from the Web and then generate the final definitions by incorporating the extracted definitional information. Our framework is remarkably simple but effective: experiments demonstrate our method can generate high-quality definitions for jargon and outperform state-of-the-art models significantly, e.g., BLEU score from 8.76 to 22.66 and human-annotated score from 2.34 to 4.04.

ProsocialDialog: A Prosocial Backbone for Conversational Agents
Hyunwoo Kim | Youngjae Yu | Liwei Jiang | Ximing Lu | Daniel Khashabi | Gunhee Kim | Yejin Choi | Maarten Sap

Most existing dialogue systems fail to respond properly to potentially unsafe user utterances by either ignoring or passively agreeing with them. To address this issue, we introduce ProsocialDialog, the first large-scale multi-turn dialogue dataset to teach conversational agents to respond to problematic content following social norms. Covering diverse unethical, problematic, biased, and toxic situations, ProsocialDialog contains responses that encourage prosocial behavior, grounded in commonsense social rules (i.e., rules-of-thumb, RoTs). Created via a human-AI collaborative framework, ProsocialDialog consists of 58K dialogues, with 331K utterances, 160K unique RoTs, and 497K dialogue safety labels accompanied by free-form rationales.With this dataset, we introduce a dialogue safety detection module, Canary, capable of generating RoTs given conversational context, and a socially-informed dialogue agent, Prost. Empirical results show that Prost generates more socially acceptable dialogues compared to other state-of-the-art language and dialogue models in both in-domain and out-of-domain settings. Additionally, Canary effectively guides conversational agents and off-the-shelf language models to generate significantly more prosocial responses. Our work highlights the promise and importance of creating and steering conversational AI to be socially responsible.

Exploiting Global and Local Hierarchies for Hierarchical Text Classification
Ting Jiang | Deqing Wang | Leilei Sun | Zhongzhi Chen | Fuzhen Zhuang | Qinghong Yang

Hierarchical text classification aims to leverage label hierarchy in multi-label text classification. Existing methods encode label hierarchy in a global view, where label hierarchy is treated as the static hierarchical structure containing all labels. Since global hierarchy is static and irrelevant to text samples, it makes these methods hard to exploit hierarchical information. Contrary to global hierarchy, local hierarchy as a structured labels hierarchy corresponding to each text sample. It is dynamic and relevant to text samples, which is ignored in previous methods. To exploit global and local hierarchies, we propose Hierarchy-guided BERT with Global and Local hierarchies (HBGL), which utilizes the large-scale parameters and prior language knowledge of BERT to model both global and local hierarchies. Moreover, HBGL avoids the intentional fusion of semantic and hierarchical modules by directly modeling semantic and hierarchical information with BERT. Compared with the state-of-the-art method HGCLR, our method achieves significant improvement on three benchmark datasets.

Semantic-aware Contrastive Learning for More Accurate Semantic Parsing
Shan Wu | Chunlei Xin | Bo Chen | Xianpei Han | Le Sun

Since the meaning representations are detailed and accurate annotations which express fine-grained sequence-level semtantics, it is usually hard to train discriminative semantic parsers via Maximum Likelihood Estimation (MLE) in an autoregressive fashion. In this paper, we propose a semantic-aware contrastive learning algorithm, which can learn to distinguish fine-grained meaning representations and take the overall sequence-level semantic into consideration. Specifically, a multi-level online sampling algorithm is proposed to sample confusing and diverse instances. Three semantic-aware similarity functions are designed to accurately measure the distance between meaning representations as a whole. And a ranked contrastive loss is proposed to pull the representations of the semantic-identical instances together and push negative instances away. Experiments on two standard datasets show that our approach achieves significant improvements over MLE baselines and gets state-of-the-art performances by simply applying semantic-aware contrastive learning on a vanilla Seq2Seq model.

Scientific Paper Extractive Summarization Enhanced by Citation Graphs
Xiuying Chen | Mingzhe Li | Shen Gao | Rui Yan | Xin Gao | Xiangliang Zhang

In a citation graph, adjacent paper nodes share related scientific terms and topics. The graph thus conveys unique structure information of document-level relatedness that can be utilized in the paper summarization task, for exploring beyond the intra-document information.In this work, we focus on leveraging citation graphs to improve scientific paper extractive summarization under different settings.We first propose a Multi-granularity Unsupervised Summarization model (MUS) as a simple and low-cost solution to the task.MUS finetunes a pre-trained encoder model on the citation graph by link prediction tasks.Then, the abstract sentences are extracted from the corresponding paper considering multi-granularity information.Preliminary results demonstrate that citation graph is helpful even in a simple unsupervised framework.Motivated by this, we next propose a Graph-based Supervised Summarizationmodel (GSS) to achieve more accurate results on the task when large-scale labeled data are available.Apart from employing the link prediction as an auxiliary task, GSS introduces a gated sentence encoder and a graph information fusion module to take advantage of the graph information to polish the sentence representation.Experiments on a public benchmark dataset show that MUS and GSS bring substantial improvements over the prior state-of-the-art model.

Hardness-guided domain adaptation to recognise biomedical named entities under low-resource scenarios
Ngoc Dang Nguyen | Lan Du | Wray Buntine | Changyou Chen | Richard Beare

Domain adaptation is an effective solution to data scarcity in low-resource scenarios. However, when applied to token-level tasks such as bioNER, domain adaptation methods often suffer from the challenging linguistic characteristics that clinical narratives possess, which leads to unsatsifactory performance. In this paper, we present a simple yet effective hardness-guided domain adaptation framework for bioNER tasks that can effectively leverage the domain hardness information to improve the adaptability of the learnt model in the low-resource scenarios. Experimental results on biomedical datasets show that our model can achieve significant performance improvement over the recently published state-of-the-art (SOTA) MetaNER model.

Syntactic Multi-view Learning for Open Information Extraction
Kuicai Dong | Aixin Sun | Jung-Jae Kim | Xiaoli Li

Open Information Extraction (OpenIE) aims to extract relational tuples from open-domain sentences. Traditional rule-based or statistical models were developed based on syntactic structure of sentence, identified by syntactic parsers. However, previous neural OpenIE models under-explored the useful syntactic information. In this paper, we model both constituency and dependency trees into word-level graphs, and enable neural OpenIE to learn from the syntactic structures. To better fuse heterogeneous information from the two graphs, we adopt multi-view learning to capture multiple relationships from them. Finally, the finetuned constituency and dependency representations are aggregated with sentential semantic representations for tuple generation. Experiments show that both constituency and dependency information, and the multi-view learning are effective.

TRIPS: Efficient Vision-and-Language Pre-training with Text-Relevant Image Patch Selection
Chaoya Jiang | Haiyang Xu | Chenliang Li | Ming Yan | Wei Ye | Shikun Zhang | Bin Bi | Songfang Huang

Vision Transformers (ViTs) have been widely used in large-scale Vision and Language Pre-training (VLP) models. Though previous VLP works have proved the effectiveness of ViTs, they still suffer from computational efficiency brought by the long visual sequence. To tackle this problem, in this paper, we propose an efficient vision-and-language pre-training model with Text-Relevant Image Patch Selection, namely TRIPS, which reduces the visual sequence progressively with a text-guided patch-selection layer in the visual backbone for efficient training and inference. The patch-selection layer can dynamically compute text-dependent visual attention to identify the attentive image tokens with text guidance and fuse inattentive ones in an end-to-end manner. Meanwhile, TRIPS does not introduce extra parameters to ViTs. Experimental results on a variety of popular benchmark datasets demonstrate that TRIPS gain a speedup of 40% over previous similar VLP models, yet with competitive or better downstream task performance.

CGoDial: A Large-Scale Benchmark for Chinese Goal-oriented Dialog Evaluation
Yinpei Dai | Wanwei He | Bowen Li | Yuchuan Wu | Zheng Cao | Zhongqi An | Jian Sun | Yongbin Li

Practical dialog systems need to deal with various knowledge sources, noisy user expressions, and the shortage of annotated data. To better solve the above problems, we propose CGoDial, a new challenging and comprehensive Chinese benchmark for multi-domain Goal-oriented Dialog evaluation. It contains 96,763 dialog sessions, and 574,949 dialog turns totally, covering three datasets with different knowledge sources: 1) a slot-based dialog (SBD) dataset with table-formed knowledge, 2) a flow-based dialog (FBD) dataset with tree-formed knowledge, and a retrieval-based dialog (RBD) dataset with candidate-formed knowledge. To bridge the gap between academic benchmarks and spoken dialog scenarios, we either collect data from real conversations or add spoken features to existing datasets via crowd-sourcing. The proposed experimental settings include the combinations of training with either the entire training set or a few-shot training set, and testing with either the standard test set or a hard test subset, which can assess model capabilities in terms of general prediction, fast adaptability and reliable robustness.

Kernel-Whitening: Overcome Dataset Bias with Isotropic Sentence Embedding
SongYang Gao | Shihan Dou | Qi Zhang | Xuanjing Huang

Dataset bias has attracted increasing attention recently for its detrimental effect on the generalization ability of fine-tuned models. The current mainstream solution is designing an additional shallow model to pre-identify biased instances. However, such two-stage methods scale up the computational complexity of training process and obstruct valid feature information while mitigating bias.To address this issue, we utilize the representation normalization method which aims at disentangling the correlations between features of encoded sentences. We find it also promising in eliminating the bias problem by providing isotropic data distribution. We further propose Kernel-Whitening, a Nystrom kernel approximation method to achieve more thorough debiasing on nonlinear spurious correlations. Our framework is end-to-end with similar time consumption to fine-tuning. Experiments show that Kernel-Whitening significantly improves the performance of BERT on out-of-distribution datasets while maintaining in-distribution accuracy.

A Unified Positive-Unlabeled Learning Framework for Document-Level Relation Extraction with Different Levels of Labeling
Ye Wang | Xinxin Liu | Wenxin Hu | Tao Zhang

Document-level relation extraction (RE) aims to identify relations between entities across multiple sentences. Most previous methods focused on document-level RE under full supervision. However, in real-world scenario, it is expensive and difficult to completely label all relations in a document because the number of entity pairs in document-level RE grows quadratically with the number of entities. To solve the common incomplete labeling problem, we propose a unified positive-unlabeled learning framework - shift and squared ranking loss positive-unlabeled (SSR-PU) learning. We use positive-unlabeled (PU) learning on document-level RE for the first time. Considering that labeled data of a dataset may lead to prior shift of unlabeled data, we introduce a PU learning under prior shift of training data. Also, using none-class score as an adaptive threshold, we propose squared ranking loss and prove its Bayesian consistency with multi-label ranking metrics. Extensive experiments demonstrate that our method achieves an improvement of about 14 F1 points relative to the previous baseline with incomplete labeling. In addition, it outperforms previous state-of-the-art results under both fully supervised and extremely unlabeled settings as well.

Automatic Generation of Socratic Subquestions for Teaching Math Word Problems
Kumar Shridhar | Jakub Macina | Mennatallah El-Assady | Tanmay Sinha | Manu Kapur | Mrinmaya Sachan

Socratic questioning is an educational method that allows students to discover answers to complex problems by asking them a series of thoughtful questions. Generation of didactically sound questions is challenging, requiring understanding of the reasoning process involved in the problem. We hypothesize that such questioning strategy can not only enhance the human performance, but also assist the math word problem (MWP) solvers.In this work, we explore the ability of large language models (LMs) in generating sequential questions for guiding math word problem-solving. We propose various guided question generation schemes based on input conditioning and reinforcement learning.On both automatic and human quality evaluations, we find that LMs constrained with desirable question properties generate superior questions and improve the overall performance of a math word problem solver. We conduct a preliminary user study to examine the potential value of such question generation models in the education domain. Results suggest that the difficulty level of problems plays an important role in determining whether questioning improves or hinders human performance. We discuss the future of using such questioning strategies in education.

Mixture of Attention Heads: Selecting Attention Heads Per Token
Xiaofeng Zhang | Yikang Shen | Zeyu Huang | Jie Zhou | Wenge Rong | Zhang Xiong

Mixture-of-Experts (MoE) networks have been proposed as an efficient way to scale up model capacity and implement conditional computing. However, the study of MoE components mostly focused on the feedforward layer in Transformer architecture. This paper proposes the Mixture of Attention Heads (MoA), a new architecture that combines multi-head attention with the MoE mechanism. MoA includes a set of attention heads that each has its own set of parameters. Given an input, a router dynamically selects a subset of k attention heads per token. This conditional computation schema allows MoA to achieve stronger performance than the standard multi-head attention layer. Furthermore, the sparsely gated MoA can easily scale up the number of attention heads and the number of parameters while preserving computational efficiency. Despite performance improvements, MoA also automatically differentiates heads’ utilities, providing a new perspective to discuss the model’s interpretability. We conducted experiments on several important tasks, including Machine Translation and Masked Language Modeling. Experiments have shown promising results on several tasks against strong baselines that involve large and very deep models.

The Optimal BERT Surgeon: Scalable and Accurate Second-Order Pruning for Large Language Models
Eldar Kurtic | Daniel Campos | Tuan Nguyen | Elias Frantar | Mark Kurtz | Benjamin Fineran | Michael Goin | Dan Alistarh

In this paper, we consider the problem of sparsifying BERT models, which are a key building block for natural language processing, in order to reduce their storage and computational cost. We introduce the Optimal BERT Surgeon (oBERT), an efficient and accurate pruning method based on approximate second-order information, which we show to yield state-of-the-art results in both stages of language tasks: pre-training and fine-tuning. Specifically, oBERT extends existing work on second-order pruning by allowing for pruning weight blocks, and is the first such method that is applicable at BERT scale. Second, we investigate compounding compression approaches to obtain highly compressed but accurate models for deployment on edge devices. These models significantly push boundaries of the current state-of-the-art sparse BERT models with respect to all metrics: model size, inference speed and task accuracy. For example, relative to the dense BERT-base, we obtain 10x model size compression with < 1% accuracy drop, 10x CPU-inference speedup with < 2% accuracy drop, and 29x CPU-inference speedup with < 7.5% accuracy drop. Our code, fully integrated with Transformers and SparseML, is available at

Information-Theoretic Text Hallucination Reduction for Video-grounded Dialogue
Sunjae Yoon | Eunseop Yoon | Hee Suk Yoon | Junyeong Kim | Chang Yoo

Video-grounded Dialogue (VGD) aims to decode an answer sentence to a question regarding a given video and dialogue context. Despite the recent success of multi-modal reasoning to generate answer sentences, existing dialogue systems still suffer from a text hallucination problem, which denotes indiscriminate text-copying from input texts without an understanding of the question. This is due to learning spurious correlations from the fact that answer sentences in the dataset usually include the words of input texts, thus the VGD system excessively relies on copying words from input texts by hoping those words to overlap with ground-truth texts. Hence, we design Text Hallucination Mitigating (THAM) framework, which incorporates Text Hallucination Regularization (THR) loss derived from the proposed information-theoretic text hallucination measurement approach. Applying THAM with current dialogue systems validates the effectiveness on VGD benchmarks (i.e., AVSD@DSTC7 and AVSD@DSTC8) and shows enhanced interpretability.

DSM: Question Generation over Knowledge Base via Modeling Diverse Subgraphs with Meta-learner
Shasha Guo | Jing Zhang | Yanling Wang | Qianyi Zhang | Cuiping Li | Hong Chen

Existing methods on knowledge base question generation (KBQG) learn a one-size-fits-all model by training together all subgraphs without distinguishing the diverse semantics of subgraphs. In this work, we show that making use of the past experience on semantically similar subgraphs can reduce the learning difficulty and promote the performance of KBQG models. To achieve this, we propose a novel approach to model diverse subgraphs with meta-learner (DSM). Specifically, we devise a graph contrastive learning-based retriever to identify semantically similar subgraphs, so that we can construct the semantics-aware learning tasks for the meta-learner to learn semantics-specific and semantics-agnostic knowledge on and across these tasks. Extensive experiments on two widely-adopted benchmarks for KBQG show that DSM derives new state-of-the-art performance and benefits the question answering tasks as a means of data augmentation.

RelU-Net: Syntax-aware Graph U-Net for Relational Triple Extraction
Yunqi Zhang | Yubo Chen | Yongfeng Huang

Relational triple extraction is a critical task for natural language processing. Existing methods mainly focused on capturing semantic information, but suffered from ignoring the syntactic structure of the sentence, which is proved in the relation classification task to contain rich relational information. This is due to the absence of entity locations, which is the prerequisite for pruning noisy edges from the dependency tree, when extracting relational triples. In this paper, we propose a unified framework to tackle this challenge and incorporate syntactic information for relational triple extraction. First, we propose to automatically contract the dependency tree into a core relational topology and eliminate redundant information with graph pooling operations. Then, we propose a symmetrical expanding path with graph unpooling operations to fuse the contracted core syntactic interactions with the original sentence context. We also propose a bipartite graph matching objective function to capture the reflections between the core topology and golden relational facts. Since our model shares similar contracting and expanding paths with encoder-decoder models like U-Net, we name our model as Relation U-Net (RelU-Net). We conduct experiments on several datasets and the results prove the effectiveness of our method.

Evidence > Intuition: Transferability Estimation for Encoder Selection
Elisa Bassignana | Max Müller-Eberstein | Mike Zhang | Barbara Plank

With the increase in availability of large pre-trained language models (LMs) in Natural Language Processing (NLP), it becomes critical to assess their fit for a specific target task a priori—as fine-tuning the entire space of available LMs is computationally prohibitive and unsustainable. However, encoder transferability estimation has received little to no attention in NLP. In this paper, we propose to generate quantitative evidence to predict which LM, out of a pool of models, will perform best on a target task without having to fine-tune all candidates. We provide a comprehensive study on LM ranking for 10 NLP tasks spanning the two fundamental problem types of classification and structured prediction. We adopt the state-of-the-art Logarithm of Maximum Evidence (LogME) measure from Computer Vision (CV) and find that it positively correlates with final LM performance in 94% of the setups.In the first study of its kind, we further compare transferability measures with the de facto standard of human practitioner ranking, finding that evidence from quantitative metrics is more robust than pure intuition and can help identify unexpected LM candidates.

Chunk-based Nearest Neighbor Machine Translation
Pedro Henrique Martins | Zita Marinho | André F. T. Martins

Semi-parametric models, which augment generation with retrieval, have led to impressive results in language modeling and machine translation, due to their ability to retrieve fine-grained information from a datastore of examples. One of the most prominent approaches, kNN-MT, exhibits strong domain adaptation capabilities by retrieving tokens from domain-specific datastores (Khandelwal et al., 2021). However, kNN-MT requires an expensive retrieval operation for every single generated token, leading to a very low decoding speed (around 8 times slower than a parametric model). In this paper, we introduce a chunk-based kNN-MT model which retrieves chunks of tokens from the datastore, instead of a single token. We propose several strategies for incorporating the retrieved chunks into the generation process, and for selecting the steps at which the model needs to search for neighbors in the datastore. Experiments on machine translation in two settings, static and “on-the-fly” domain adaptation, show that the chunk-based kNN-MT model leads to significant speed-ups (up to 4 times) with only a small drop in translation quality.

FiE: Building a Global Probability Space by Leveraging Early Fusion in Encoder for Open-Domain Question Answering
Akhil Kedia | Mohd Abbas Zaidi | Haejun Lee

Generative models have recently started to outperform extractive models in Open Domain Question Answering, largely by leveraging their decoder to attend over multiple encoded passages and combining their information. However, generative models tend to be larger than extractive models due to the need for a decoder, run slower during inference due to auto-regressive decoder beam search, and their generated output often suffers from hallucinations. We propose to extend transformer encoders with the ability to fuse information from multiple passages, using global representation to provide cross-sample attention over all tokens across samples. Furthermore, we propose an alternative answer span probability calculation to better aggregate answer scores in the global space of all samples. Using our proposed method, we outperform the current state-of-the-art method by 2.5 Exact Match score on the Natural Question dataset while using only 25% of parameters and 35% of the latency during inference, and 4.4 Exact Match on WebQuestions dataset. When coupled with synthetic data augmentation, we outperform larger models on the TriviaQA dataset as well. The latency and parameter savings of our method make it particularly attractive for open-domain question answering, as these models are often compute-intensive.

Inductive Relation Prediction with Logical Reasoning Using Contrastive Representations
Yudai Pan | Jun Liu | Lingling Zhang | Tianzhe Zhao | Qika Lin | Xin Hu | Qianying Wang

Relation prediction in knowledge graphs (KGs) aims at predicting missing relations in incomplete triples, whereas the dominant embedding paradigm has a restriction on handling unseen entities during testing. In the real-world scenario, the inductive setting is more common because entities in the training process are finite. Previous methods capture an inductive ability by implicit logic in KGs. However, it would be challenging to preciously acquire entity-independent relational semantics of compositional logic rules and to deal with the deficient supervision of logic caused by the scarcity of relational semantics. To this end, we propose a novel graph convolutional network (GCN)-based model LogCo with logical reasoning by contrastive representations. LogCo firstly extracts enclosing subgraphs and relational paths between two entities to supply the entity-independence. Then a contrastive strategy for relational path instances and the subgraph is proposed for the issue of deficient supervision. The contrastive representations are learned for a joint training regime. Finally, prediction results and logic rules for reasoning are attained. Comprehensive experiments on twelve inductive datasets show that LogCo achieves outstanding performance comparing with state-of-the-art inductive relation prediction baselines.

Improving Chinese Spelling Check by Character Pronunciation Prediction: The Effects of Adaptivity and Granularity
Jiahao Li | Quan Wang | Zhendong Mao | Junbo Guo | Yanyan Yang | Yongdong Zhang

Chinese spelling check (CSC) is a fundamental NLP task that detects and corrects spelling errors in Chinese texts. As most of these spelling errors are caused by phonetic similarity, effectively modeling the pronunciation of Chinese characters is a key factor for CSC. In this paper, we consider introducing an auxiliary task of Chinese pronunciation prediction (CPP) to improve CSC, and, for the first time, systematically discuss the adaptivity and granularity of this auxiliary task. We propose SCOPE which builds upon a shared encoder two parallel decoders, one for the primary CSC task and the other for a fine-grained auxiliary CPP task, with a novel adaptive weighting scheme to balance the two tasks. In addition, we design a delicate iterative correction strategy for further improvements during inference. Empirical evaluation shows that SCOPE achieves new state-of-the-art on three CSC benchmarks, demonstrating the effectiveness and superiority of the auxiliary CPP task. Comprehensive ablation studies further verify the positive effects of adaptivity and granularity of the task.

MT-GenEval: A Counterfactual and Contextual Dataset for Evaluating Gender Accuracy in Machine Translation
Anna Currey | Maria Nadejde | Raghavendra Reddy Pappagari | Mia Mayer | Stanislas Lauly | Xing Niu | Benjamin Hsu | Georgiana Dinu

As generic machine translation (MT) quality has improved, the need for targeted benchmarks that explore fine-grained aspects of quality has increased. In particular, gender accuracy in translation can have implications in terms of output fluency, translation accuracy, and ethics. In this paper, we introduce MT-GenEval, a benchmark for evaluating gender accuracy in translation from English into eight widely-spoken languages. MT-GenEval complements existing benchmarks by providing realistic, gender-balanced, counterfactual data in eight language pairs where the gender of individuals is unambiguous in the input segment, including multi-sentence segments requiring inter-sentential gender agreement. Our data and code is publicly available under a CC BY SA 3.0 license.

A Span-level Bidirectional Network for Aspect Sentiment Triplet Extraction
Yuqi Chen | Chen Keming | Xian Sun | Zequn Zhang

Aspect Sentiment Triplet Extraction (ASTE) is a new fine-grained sentiment analysis task that aims to extract triplets of aspect terms, sentiments, and opinion terms from review sentences. Recently, span-level models achieve gratifying results on ASTE task by taking advantage of the predictions of all possible spans. Since all possible spans significantly increases the number of potential aspect and opinion candidates, it is crucial and challenging to efficiently extract the triplet elements among them. In this paper, we present a span-level bidirectional network which utilizes all possible spans as input and extracts triplets from spans bidirectionally. Specifically, we devise both the aspect decoder and opinion decoder to decode the span representations and extract triples from aspect-to-opinion and opinion-to-aspect directions. With these two decoders complementing with each other, the whole network can extract triplets from spans more comprehensively. Moreover, considering that mutual exclusion cannot be guaranteed between the spans, we design a similar span separation loss to facilitate the downstream task of distinguishing the correct span by expanding the KL divergence of similar spans during the training process; in the inference process, we adopt an inference strategy to remove conflicting triplets from the results base on their confidence scores. Experimental results show that our framework not only significantly outperforms state-of-the-art methods, but achieves better performance in predicting triplets with multi-token entities and extracting triplets in sentences contain multi-triplets.

On the Calibration of Massively Multilingual Language Models
Kabir Ahuja | Sunayana Sitaram | Sandipan Dandapat | Monojit Choudhury

Massively Multilingual Language Models (MMLMs) have recently gained popularity due to their surprising effectiveness in cross-lingual transfer. While there has been much work in evaluating these models for their performance on a variety of tasks and languages, little attention has been paid on how well calibrated these models are with respect to the confidence in their predictions. We first investigate the calibration of MMLMs in the zero-shot setting and observe a clear case of miscalibration in low-resource languages or those which are typologically diverse from English. Next, we empirically show that calibration methods like temperature scaling and label smoothing do reasonably well in improving calibration in the zero-shot scenario. We also find that few-shot examples in the language can further help reduce calibration errors, often substantially. Overall, our work contributes towards building more reliable multilingual models by highlighting the issue of their miscalibration, understanding what language and model-specific factors influence it, and pointing out the strategies to improve the same.

Momentum Contrastive Pre-training for Question Answering
Minda Hu | Muzhi Li | Yasheng Wang | Irwin King

Existing pre-training methods for extractive Question Answering (QA) generate cloze-like queries different from natural questions in syntax structure, which could overfit pre-trained models to simple keyword matching. In order to address this problem, we propose a novel Momentum Contrastive pRe-training fOr queStion anSwering (MCROSS) method for extractive QA. Specifically, MCROSS introduces a momentum contrastive learning framework to align the answer probability between cloze-like and natural query-passage sample pairs. Hence, the pre-trained models can better transfer the knowledge learned in cloze-like samples to answering natural questions. Experimental results on three benchmarking QA datasets show that our method achieves noticeable improvement compared with all baselines in both supervised and zero-shot scenarios.

A Second Wave of UD Hebrew Treebanking and Cross-Domain Parsing
Amir Zeldes | Nick Howell | Noam Ordan | Yifat Ben Moshe

Foundational Hebrew NLP tasks such as segmentation, tagging and parsing, have relied to date on various versions of the Hebrew Treebank (HTB, Sima’an et al. 2001). However, the data in HTB, a single-source newswire corpus, is now over 30 years old, and does not cover many aspects of contemporary Hebrew on the web. This paper presents a new, freely available UD treebank of Hebrew stratified from a range of topics selected from Hebrew Wikipedia. In addition to introducing the corpus and evaluating the quality of its annotations, we deploy automatic validation tools based on grew (Guillaume, 2021), and conduct the first cross domain parsing experiments in Hebrew. We obtain new state-of-the-art (SOTA) results on UD NLP tasks, using a combination of the latest language modelling and some incremental improvements to existing transformer based approaches. We also release a new version of the UD HTB matching annotation scheme updates from our new corpus.

Finding Dataset Shortcuts with Grammar Induction
Dan Friedman | Alexander Wettig | Danqi Chen

Many NLP datasets have been found to contain shortcuts: simple decision rules that achieve surprisingly high accuracy. However, it is difficult to discover shortcuts automatically. Prior work on automatic shortcut detection has focused on enumerating features like unigrams or bigrams, which can find only low-level shortcuts, or relied on post-hoc model interpretability methods like saliency maps, which reveal qualitative patterns without a clear statistical interpretation. In this work, we propose to use probabilistic grammars to characterize and discover shortcuts in NLP datasets. Specifically, we use a context-free grammar to model patterns in sentence classification datasets and use a synchronous context-free grammar to model datasets involving sentence pairs. The resulting grammars reveal interesting shortcut features in a number of datasets, including both simple and high-level features, and automatically identify groups of test examples on which conventional classifiers fail. Finally, we show that the features we discover can be used to generate diagnostic contrast examples and incorporated into standard robust optimization methods to improve worst-group accuracy.

Retrieval Augmentation for Commonsense Reasoning: A Unified Approach
Wenhao Yu | Chenguang Zhu | Zhihan Zhang | Shuohang Wang | Zhuosheng Zhang | Yuwei Fang | Meng Jiang

A common thread of retrieval-augmented methods in the existing literature focuses on retrieving encyclopedic knowledge, such as Wikipedia, which facilitates well-defined entity and relation spaces that can be modeled. However, applying such methods to commonsense reasoning tasks faces two unique challenges, i.e., the lack of a general large-scale corpus for retrieval and a corresponding effective commonsense retriever. In this paper, we systematically investigate how to leverage commonsense knowledge retrieval to improve commonsense reasoning tasks. We proposed a unified framework of retrieval-augmented commonsense reasoning (called RACo), including a newly constructed commonsense corpus with over 20 million documents and novel strategies for training a commonsense retriever. We conducted experiments on four different commonsense reasoning tasks. Extensive evaluation results showed that our proposed RACo can significantly outperform other knowledge-enhanced method counterparts, achieving new SoTA performance on the CommonGen and CREAK leaderboards.

Open World Classification with Adaptive Negative Samples
Ke Bai | Guoyin Wang | Jiwei Li | Sunghyun Park | Sungjin Lee | Puyang Xu | Ricardo Henao | Lawrence Carin

Open world classification is a task in natural language processing with key practical relevance and impact.Since the open or unknown category data only manifests in the inference phase, finding a model with a suitable decision boundary accommodating for the identification of known classes and discrimination of the open category is challenging.The performance of existing models is limited by the lack of effective open category data during the training stage or the lack of a good mechanism to learn appropriate decision boundaries.We propose an approach based on Adaptive Negative Samples (ANS) designed to generate effective synthetic open category samples in the training stage and without requiring any prior knowledge or external datasets.Empirically, we find a significant advantage in using auxiliary one-versus-rest binary classifiers, which effectively utilize the generated negative samples and avoid the complex threshold-seeking stage in previous works.Extensive experiments on three benchmark datasets show that ANS achieves significant improvements over state-of-the-art methods.

Re3: Generating Longer Stories With Recursive Reprompting and Revision
Kevin Yang | Yuandong Tian | Nanyun Peng | Dan Klein

We consider the problem of automatically generating longer stories of over two thousand words. Compared to prior work on shorter stories, long-range plot coherence and relevance are more central challenges here. We propose the Recursive Reprompting and Revision framework (Re3) to address these challenges by (a) prompting a general-purpose language model to construct a structured overarching plan, and (b) generating story passages by repeatedly injecting contextual information from both the plan and current story state into a language model prompt. We then revise by (c) reranking different continuations for plot coherence and premise relevance, and finally (d) editing the best continuation for factual consistency. Compared to similar-length stories generated directly from the same base model, human evaluators judged substantially more of Re3’s stories as having a coherent overarching plot (by 14% absolute increase), and relevant to the given initial premise (by 20%).

Does Joint Training Really Help Cascaded Speech Translation?
Viet Anh Khoa Tran | David Thulke | Yingbo Gao | Christian Herold | Hermann Ney

Currently, in speech translation, the straightforward approach - cascading a recognition system with a translation system - delivers state-of-the-art results.However, fundamental challenges such as error propagation from the automatic speech recognition system still remain.To mitigate these problems, recently, people turn their attention to direct data and propose various joint training methods.In this work, we seek to answer the question of whether joint training really helps cascaded speech translation.We review recent papers on the topic and also investigate a joint training criterion by marginalizing the transcription posterior probabilities.Our findings show that a strong cascaded baseline can diminish any improvements obtained using joint training, and we suggest alternatives to joint training.We hope this work can serve as a refresher of the current speech translation landscape, and motivate research in finding more efficient and creative ways to utilize the direct data for speech translation.

MasakhaNER 2.0: Africa-centric Transfer Learning for Named Entity Recognition
David Adelani | Graham Neubig | Sebastian Ruder | Shruti Rijhwani | Michael Beukman | Chester Palen-Michel | Constantine Lignos | Jesujoba Alabi | Shamsuddeen Muhammad | Peter Nabende | Cheikh M. Bamba Dione | Andiswa Bukula | Rooweither Mabuya | Bonaventure F. P. Dossou | Blessing Sibanda | Happy Buzaaba | Jonathan Mukiibi | Godson Kalipe | Derguene Mbaye | Amelia Taylor | Fatoumata Kabore | Chris Chinenye Emezue | Anuoluwapo Aremu | Perez Ogayo | Catherine Gitau | Edwin Munkoh-Buabeng | Victoire Memdjokam Koagne | Allahsera Auguste Tapo | Tebogo Macucwa | Vukosi Marivate | Mboning Tchiaze Elvis | Tajuddeen Gwadabe | Tosin Adewumi | Orevaoghene Ahia | Joyce Nakatumba-Nabende | Neo Lerato Mokono | Ignatius Ezeani | Chiamaka Chukwuneke | Mofetoluwa Oluwaseun Adeyemi | Gilles Quentin Hacheme | Idris Abdulmumin | Odunayo Ogundepo | Oreen Yousuf | Tatiana Moteu | Dietrich Klakow

African languages are spoken by over a billion people, but they are under-represented in NLP research and development. Multiple challenges exist, including the limited availability of annotated training and evaluation datasets as well as the lack of understanding of which settings, languages, and recently proposed methods like cross-lingual transfer will be effective. In this paper, we aim to move towards solutions for these challenges, focusing on the task of named entity recognition (NER). We present the creation of the largest to-date human-annotated NER dataset for 20 African languages. We study the behaviour of state-of-the-art cross-lingual transfer methods in an Africa-centric setting, empirically demonstrating that the choice of source transfer language significantly affects performance. While much previous work defaults to using English as the source language, our results show that choosing the best transfer language improves zero-shot F1 scores by an average of 14% over 20 languages as compared to using English.

Ethics consideration sections in natural language processing papers
Luciana Benotti | Patrick Blackburn

In this paper, we present the results of a manual classification of all ethical consideration sections for ACL 2021. We also compare how many papers had an ethics consideration section per track and per world region in ACL 2021. We classified papers according to the ethical issues covered (research benefits, potential harms, and vulnerable groups affected) and whether the paper was marked as requiring ethics review by at least one reviewer. Moreover, we discuss recurring obstacles we have observed (highlighting some interesting texts we found along the way) and conclude with three suggestions. We think that this paper may be useful for anyone who needs to write — or review — an ethics section and would like to get an overview of what others have done.

Continued Pretraining for Better Zero- and Few-Shot Promptability
Zhaofeng Wu | Robert L Logan IV | Pete Walsh | Akshita Bhagia | Dirk Groeneveld | Sameer Singh | Iz Beltagy

Recently introduced language model prompting methods can achieve high accuracy in zero- and few-shot settings while requiring few to no learned task-specific parameters. Nevertheless, these methods still often trail behind full model finetuning. In this work, we investigate if a dedicated continued pretraining stage could improve “promptability”, i.e., zero-shot performance with natural language prompts or few-shot performance with prompt tuning. We reveal settings where existing continued pretraining methods lack promptability. We also identify current methodological gaps, which we fill with thorough large-scale experiments. We demonstrate that a simple recipe, continued pretraining that incorporates a trainable prompt during multi-task learning, leads to improved promptability in both zero- and few-shot settings compared to existing methods, up to 31% relative. On the other hand, we find that continued pretraining using MAML-style meta-learning, a method that directly optimizes few-shot promptability, yields subpar performance. We validate our findings with two prompt tuning methods, and, based on our results, we provide concrete recommendations to optimize promptability for different use cases.

Less is More: Summary of Long Instructions is Better for Program Synthesis
Kirby Kuznia | Swaroop Mishra | Mihir Parmar | Chitta Baral

Despite the success of large pre-trained language models (LMs) such as Codex, they show below-par performance on the larger and more complicated programming related questions. We show that LMs benefit from the summarized version of complicated questions. Our findings show that superfluous information often present in problem description such as human characters, background stories, and names (which are included to help humans in understanding a task) does not help models in understanding a task. To this extent, we create a meta-dataset from the frequently used APPS dataset and the newly created CodeContests dataset for the program synthesis task. Our meta-dataset consists of human and synthesized summaries of the long and complicated programming questions. Experimental results on Codex show that our proposed approach outperforms baseline by 8.13% on the APPS dataset and 11.88% on the CodeContests dataset on an average in terms of strict accuracy. Our analysis shows that summaries significantly improve performance for introductory (9.86%) and interview (11.48%) related programming questions. However, it shows improvement by a small margin ( 2%) for competitive programming questions, implying the scope for future research direction.

Is a Question Decomposition Unit All We Need?
Pruthvi Patel | Swaroop Mishra | Mihir Parmar | Chitta Baral

Large Language Models (LMs) have achieved state-of-the-art performance on many Natural Language Processing (NLP) benchmarks. With the growing number of new benchmarks, we build bigger and more complex LMs. However, building new LMs may not be an ideal option owing to the cost, time and environmental impact associated with it. We explore an alternative route: can we modify data by expressing it in terms of the model’s strengths, so that a question becomes easier for models to answer? We investigate if humans can decompose a hard question into a set of simpler questions that are relatively easier for models to solve. We analyze a range of datasets involving various forms of reasoning and find that it is indeed possible to significantly improve model performance (24% for GPT3 and 29% for RoBERTa-SQuAD along with a symbolic calculator) via decomposition. Our approach provides a viable option to involve people in NLP research in a meaningful way. Our findings indicate that Human-in-the-loop Question Decomposition (HQD) can potentially provide an alternate path to building large LMs.

Discourse-Aware Soft Prompting for Text Generation
Marjan Ghazvininejad | Vladimir Karpukhin | Vera Gor | Asli Celikyilmaz

Current efficient fine-tuning methods(e.g., adapters, prefix-tuning, etc.) have optimized conditional text generation via training a small set of extra parameters of the neural language model, while freezing the rest for efficiency. While showing strong performance on some generation tasks, they don’t generalize across all generation tasks. We show that soft-prompt based conditional text generation can be improved with simple and efficient methods that simulate modeling the discourse structure of human written text.We investigate two design choices: First, we apply hierarchical blocking on the prefix parameters to simulate a higher-level discourse structure of human written text. Second, we apply attention sparsity on the prefix parameters at different layers of the network and learn sparse transformations on the softmax-function. We show that structured design of prefix parameters yields more coherent, faithful and relevant generations than the baseline prefix-tuning on all generation tasks.

ExPUNations: Augmenting Puns with Keywords and Explanations
Jiao Sun | Anjali Narayan-Chen | Shereen Oraby | Alessandra Cervone | Tagyoung Chung | Jing Huang | Yang Liu | Nanyun Peng

The tasks of humor understanding and generation are challenging and subjective even for humans, requiring commonsense and real-world knowledge to master. Puns, in particular, add the challenge of fusing that knowledge with the ability to interpret lexical-semantic ambiguity. In this paper, we present the ExPUNations (ExPUN) dataset, in which we augment an existing dataset of puns with detailed crowdsourced annotations of keywords denoting the most distinctive words that make the text funny, pun explanations describing why the text is funny, and fine-grained funniness ratings. This is the first humor dataset with such extensive and fine-grained annotations specifically for puns. Based on these annotations, we propose two tasks: explanation generation to aid with pun classification and keyword-conditioned pun generation, to challenge the current state-of-the-art natural language understanding and generation models’ ability to understand and generate humor. We showcase that the annotated keywords we collect are helpful for generating better novel humorous texts in human evaluation, and that our natural language explanations can be leveraged to improve both the accuracy and robustness of humor classifiers.

SLING: Sino Linguistic Evaluation of Large Language Models
Yixiao Song | Kalpesh Krishna | Rajesh Bhatt | Mohit Iyyer

To understand what kinds of linguistic knowledge are encoded by pretrained Chinese language models (LMs), we introduce the benchmark of Sino LINGuistics (SLING), which consists of 38K minimal sentence pairs in Mandarin Chinese grouped into 9 high-level linguistic phenomena. Each pair demonstrates the acceptability contrast of a specific syntactic or semantic phenomenon (e.g., The keys are lost vs. The keys is lost), and an LM should assign lower perplexity to the acceptable sentence. In contrast to the CLiMP dataset (Xiang et al., 2021), which also contains Chinese minimal pairs and was created by translating the vocabulary of the English BLiMP dataset, the minimal pairs in SLING are derived primarily by applying syntactic and lexical transformations to naturally-occurring, linguist-annotated sentences from the Chinese Treebank 9.0, thus addressing severe issues in CLiMP’s data generation process. We test 18 publicly available pretrained monolingual (e.g., BERT-base-zh, CPM) and multi-lingual (e.g., mT5, XLM) language models on SLING. Our experiments show that the average accuracy for LMs is far below human performance (69.7% vs. 97.1%), while BERT-base-zh achieves the highest accuracy (84.8%) of all tested LMs, even much larger ones. Additionally, we find that most LMs have a strong gender and number (singular/plural) bias, and they perform better on local phenomena than hierarchical ones.

Context-Situated Pun Generation
Jiao Sun | Anjali Narayan-Chen | Shereen Oraby | Shuyang Gao | Tagyoung Chung | Jing Huang | Yang Liu | Nanyun Peng

Previous work on pun generation commonly begins with a given pun word (a pair of homophones for heterographic pun generation and a polyseme for homographic pun generation) and seeks to generate an appropriate pun. While this may enable efficient pun generation, we believe that a pun is most entertaining if it fits appropriately within a given context, e.g., a given situation or dialogue. In this work, we propose a new task, context-situated pun generation, where a specific context represented by a set of keywords is provided, and the task is to first identify suitable pun words that are appropriate for the context, then generate puns based on the context keywords and the identified pun words. We collect a new dataset, CUP (Context-sitUated Pun), containing 4.5k tuples of context words and pun pairs. Based on the new data and setup, we propose a pipeline system for context-situated pun generation, including a pun word retrieval module that identifies suitable pun words for a given context, and a pun generation module that generates puns from context keywords and pun words. Human evaluation shows that 69% of our top retrieved pun words can be used to generate context-situated puns, and our generation module yields successful puns 31% of the time given a plausible tuple of context words and pun pair, almost tripling the yield of a state-of-the-art pun generation model. With an end-to-end evaluation, our pipeline system with the top-1 retrieved pun pair for a given context can generate successful puns 40% of the time, better than all other modeling variations but 32% lower than the human success rate. This highlights the difficulty of the task, and encourages more research in this direction.

Retrieval-Augmented Generative Question Answering for Event Argument Extraction
Xinya Du | Heng Ji

Event argument extraction has long been studied as a sequential prediction problem with extractive-based methods, tackling each argument in isolation. Although recent work proposes generation-based methods to capture cross-argument dependency, they require generating and post-processing a complicated target sequence (template). Motivated by these observations and recent pretrained language models’ capabilities of learning from demonstrations. We propose a retrieval-augmented generative QA model (R-GQA) for event argument extraction. It retrieves the most similar QA pair and augments it as prompt to the current example’s context, then decodes the arguments as answers. Our approach outperforms substantially prior methods across various settings (i.e. fully supervised, domain transfer, and fewshot learning). Finally, we propose a clustering-based sampling strategy (JointEnc) and conduct a thorough analysis of how different strategies influence the few-shot learning performances.

Concadia: Towards Image-Based Text Generation with a Purpose
Elisa Kreiss | Fei Fang | Noah Goodman | Christopher Potts

Current deep learning models often achieve excellent results on benchmark image-to-text datasets but fail to generate texts that are useful in practice. We argue that to close this gap, it is vital to distinguish descriptions from captions based on their distinct communicative roles. Descriptions focus on visual features and are meant to replace an image (often to increase accessibility), whereas captions appear alongside an image to supply additional information. To motivate this distinction and help people put it into practice, we introduce the publicly available Wikipedia-based dataset Concadia consisting of 96,918 images with corresponding English-language descriptions, captions, and surrounding context. Using insights from Concadia, models trained on it, and a preregistered human-subjects experiment with human- and model-generated texts, we characterize the commonalities and differences between descriptions and captions. In addition, we show that, for generating both descriptions and captions, it is useful to augment image-to-text models with representations of the textual context in which the image appeared.

Context Matters for Image Descriptions for Accessibility: Challenges for Referenceless Evaluation Metrics
Elisa Kreiss | Cynthia Bennett | Shayan Hooshmand | Eric Zelikman | Meredith Ringel Morris | Christopher Potts

Few images on the Web receive alt-text descriptions that would make them accessible to blind and low vision (BLV) users. Image-based NLG systems have progressed to the point where they can begin to address this persistent societal problem, but these systems will not be fully successful unless we evaluate them on metrics that guide their development correctly. Here, we argue against current referenceless metrics – those that don’t rely on human-generated ground-truth descriptions – on the grounds that they do not align with the needs of BLV users. The fundamental shortcoming of these metrics is that they do not take context into account, whereas contextual information is highly valued by BLV users. To substantiate these claims, we present a study with BLV participants who rated descriptions along a variety of dimensions. An in-depth analysis reveals that the lack of context-awareness makes current referenceless metrics inadequate for advancing image accessibility. As a proof-of-concept, we provide a contextual version of the referenceless metric CLIPScore which begins to address the disconnect to the BLV data.

MetaLogic: Logical Reasoning Explanations with Fine-Grained Structure
Yinya Huang | Hongming Zhang | Ruixin Hong | Xiaodan Liang | Changshui Zhang | Dong Yu

In this paper, we propose a comprehensive benchmark to investigate models’ logical reasoning capabilities in complex real-life scenarios. Current explanation datasets often employ synthetic data with simple reasoning structures. Therefore, it cannot express more complex reasoning processes, such as the rebuttal to a reasoning step and the degree of certainty of the evidence. To this end, we propose a comprehensive logical reasoning explanation form. Based on the multi-hop chain of reasoning, the explanation form includes three main components: (1) The condition of rebuttal that the reasoning node can be challenged; (2) Logical formulae that uncover the internal texture of reasoning nodes; (3) Reasoning strength indicated by degrees of certainty. The fine-grained structure conforms to the real logical reasoning scenario, better fitting the human cognitive process but, simultaneously, is more challenging for the current models. We evaluate the current best models’ performance on this new explanation form. The experimental results show that generating reasoning graphs remains a challenging task for current models, even with the help of giant pre-trained language models.

Explicit Query Rewriting for Conversational Dense Retrieval
Hongjin Qian | Zhicheng Dou

In a conversational search scenario, a query might be context-dependent because some words are referred to previous expressions or omitted. Previous works tackle the issue by either reformulating the query into a self-contained query (query rewriting) or learning a contextualized query embedding from the query context (context modelling). In this paper, we propose a model CRDR that can perform query rewriting and context modelling in a unified framework in which the query rewriting’s supervision signals further enhance the context modelling. Instead of generating a new query, CRDR only performs necessary modifications on the original query, which improves both accuracy and efficiency of query rewriting. In the meantime, the query rewriting benefits the context modelling by explicitly highlighting relevant terms in the query context, which improves the quality of the learned contextualized query embedding. To verify the effectiveness of CRDR, we perform comprehensive experiments on TREC CAsT-19 and TREC CAsT-20 datasets, and the results show that our method outperforms all baseline models in terms of both quality of query rewriting and quality of context-aware ranking.

Efficient Nearest Neighbor Emotion Classification with BERT-whitening
Wenbiao Yin | Lin Shang

Retrieval-based methods have been proven effective in many NLP tasks. Previous methods use representations from the pre-trained model for similarity search directly. However, the sentence representations from the pre-trained model like BERT perform poorly in retrieving semantically similar sentences, resulting in poor performance of the retrieval-based methods. In this paper, we propose kNN-EC, a simple and efficient non-parametric emotion classification (EC) method using nearest neighbor retrieval. We use BERT-whitening to get better sentence semantics, ensuring that nearest neighbor retrieval works. Meanwhile, BERT-whitening can also reduce memory storage of datastore and accelerate retrieval speed, solving the efficiency problem of the previous methods. kNN-EC average improves the pre-trained model by 1.17 F1-macro on two emotion classification datasets.

FastClass: A Time-Efficient Approach to Weakly-Supervised Text Classification
Tingyu Xia | Yue Wang | Yuan Tian | Yi Chang

Weakly-supervised text classification aims to train a classifier using only class descriptions and unlabeled data. Recent research shows that keyword-driven methods can achieve state-of-the-art performance on various tasks. However, these methods not only rely on carefully-crafted class descriptions to obtain class-specific keywords but also require substantial amount of unlabeled data and takes a long time to train. This paper proposes FastClass, an efficient weakly-supervised classification approach. It uses dense text representation to retrieve class-relevant documents from external unlabeled corpus and selects an optimal subset to train a classifier. Compared to keyword-driven methods, our approach is less reliant on initial class descriptions as it no longer needs to expand each class description into a set of class-specific keywords.Experiments on a wide range of classification tasks show that the proposed approach frequently outperforms keyword-driven models in terms of classification accuracy and often enjoys orders-of-magnitude faster training speed.

Neural-Symbolic Inference for Robust Autoregressive Graph Parsing via Compositional Uncertainty Quantification
Zi Lin | Jeremiah Liu | Jingbo Shang

Pre-trained seq2seq models excel at graph semantic parsing with rich annotated data, but generalize worse to out-of-distribution (OOD) and long-tail examples. In comparison, symbolic parsers under-perform on population-level metrics, but exhibit unique strength in OOD and tail generalization. In this work, we study compositionality-aware approach to neural-symbolic inference informed by model confidence, performing fine-grained neural-symbolic reasoning at subgraph level (i.e., nodes and edges) and precisely targeting subgraph components with high uncertainty in the neural parser. As a result, the method combines the distinct strength of the neural and symbolic approaches in capturing different aspects of the graph prediction, leading to well-rounded generalization performance both across domains and in the tail. We empirically investigate the approach in the English Resource Grammar (ERG) parsing problem on a diverse suite of standard in-domain and seven OOD corpora. Our approach leads to 35.26% and 35.60% error reduction in aggregated SMATCH score over neural and symbolic approaches respectively, and 14% absolute accuracy gain in key tail linguistic categories over the neural model, outperforming prior state-of-art methods that do not account for compositionality or uncertainty.

A Speaker-Aware Co-Attention Framework for Medical Dialogue Information Extraction
Yuan Xia | Zhenhui Shi | Jingbo Zhou | Jiayu Xu | Chao Lu | Yehui Yang | Lei Wang | Haifeng Huang | Xia Zhang | Junwei Liu

With the development of medical digitization, the extraction and structuring of Electronic Medical Records (EMRs) have become challenging but fundamental tasks. How to accurately and automatically extract structured information from medical dialogues is especially difficult because the information needs to be inferred from complex interactions between the doctor and the patient. To this end, in this paper, we propose a speaker-aware co-attention framework for medical dialogue information extraction. To better utilize the pre-trained language representation model to perceive the semantics of the utterance and the candidate item, we develop a speaker-aware dialogue encoder with multi-task learning, which considers the speaker’s identity into account. To deal with complex interactions between different utterances and the correlations between utterances and candidate items, we propose a co-attention fusion network to aggregate the utterance information. We evaluate our framework on the public medical dialogue extraction datasets to demonstrate the superiority of our method, which can outperform the state-of-the-art methods by a large margin. Codes will be publicly available upon acceptance.

Towards Interactivity and Interpretability: A Rationale-based Legal Judgment Prediction Framework
Yiquan Wu | Yifei Liu | Weiming Lu | Yating Zhang | Jun Feng | Changlong Sun | Fei Wu | Kun Kuang

Legal judgment prediction (LJP) is a fundamental task in legal AI, which aims to assist the judge to hear the case and determine the judgment. The legal judgment usually consists of the law article, charge, and term of penalty. In the real trial scenario, the judge usually makes the decision step-by-step: first concludes the rationale according to the case’s facts and then determines the judgment. Recently, many models have been proposed and made tremendous progress in LJP, but most of them adopt an end-to-end manner that cannot be manually intervened by the judge for practical use. Moreover, existing models lack interpretability due to the neglect of rationale in the prediction process. Following the judge’s real trial logic, in this paper, we propose a novel Rationale-based Legal Judgment Prediction (RLJP) framework. In the RLJP framework, the LJP process is split into two steps. In the first phase, the model generates the rationales according to the fact description. Then it predicts the judgment based on the fact and the generated rationales. Extensive experiments on a real-world dataset show RLJP achieves the best results compared to the state-of-the-art models. Meanwhile, the proposed framework provides good interactivity and interpretability which enables practical use.

RelCLIP: Adapting Language-Image Pretraining for Visual Relationship Detection via Relational Contrastive Learning
Yi Zhu | Zhaoqing Zhu | Bingqian Lin | Xiaodan Liang | Feng Zhao | Jianzhuang Liu

Conventional visual relationship detection models only use the numeric ids of relation labels for training, but ignore the semantic correlation between the labels, which leads to severe training biases and harms the generalization ability of representations. In this paper, we introduce compact language information of relation labels for regularizing the representation learning of visual relations. Specifically, we propose a simple yet effective visual Relationship prediction framework that transfers natural language knowledge learned from Contrastive Language-Image Pre-training (CLIP) models to enhance the relationship prediction, termed RelCLIP. Benefiting from the powerful visual-semantic alignment ability of CLIP at image level, we introduce a novel Relational Contrastive Learning (RCL) approach which explores relation-level visual-semantic alignment via learning to match cross-modal relational embeddings. By collaboratively learning the semantic coherence and discrepancy from relation triplets, the model can generate more discriminative and robust representations. Experimental results on the Visual Genome dataset show that RelCLIP achieves significant improvements over strong baselines under full (provide accurate labels) and distant supervision (provide noise labels), demonstrating its powerful generalization ability in learning relationship representations. Code will be available at

Candidate Soups: Fusing Candidate Results Improves Translation Quality for Non-Autoregressive Translation
Huanran Zheng | Wei Zhu | Pengfei Wang | Xiaoling Wang

Non-autoregressive translation (NAT) model achieves a much faster inference speed than the autoregressive translation (AT) model because it can simultaneously predict all tokens during inference. However, its translation quality suffers from degradation compared to AT. And existing NAT methods only focus on improving the NAT model’s performance but do not fully utilize it. In this paper, we propose a simple but effective method called “Candidate Soups,” which can obtain high-quality translations while maintaining the inference speed of NAT models. Unlike previous approaches that pick the individual result and discard the remainders, Candidate Soups (CDS) can fully use the valuable information in the different candidate translations through model uncertainty. Extensive experiments on two benchmarks (WMT’14 EN–DE and WMT’16 EN–RO) demonstrate the effectiveness and generality of our proposed method, which can significantly improve the translation quality of various base models. More notably, our best variant outperforms the AT model on three translation tasks with 7.6× speedup.

Evaluating Parameter Efficient Learning for Generation
Peng Xu | Mostofa Patwary | Shrimai Prabhumoye | Virginia Adams | Ryan Prenger | Wei Ping | Nayeon Lee | Mohammad Shoeybi | Bryan Catanzaro

Parameter efficient learning methods (PERMs)have recently gained significant attention asthey provide an efficient way for pre-trainedlanguage models (PLMs) to adapt to a downstream task. However, these conclusions aremostly drawn from in-domain evaluations overthe full training set. In this paper, we presentcomparisons between PERMs and finetuningfrom three new perspectives: (1) the effect ofsample and model size to in-domain evaluations, (2) generalization to unseen domains andnew datasets, and (3) the faithfulness of generations. Our results show that for in-domainsettings (a) there is a cross point of samplesize for which PERMs will perform better thanfinetuning when training with fewer samples,and (b) larger PLMs have larger cross points.For cross-domain and cross-dataset cases, weshow that (a) Adapter (Houlsby et al., 2019)performs the best amongst all the PERMs studied here, and (b) it outperforms finetuning ifthe task dataset is below a certain size. Wealso compare the faithfulness of generationsand show that PERMs can achieve better faithfulness score than finetuning, especially forsmall training set, by as much as 6%. Finally,we apply Adapter to MT-NLG 530b (Smithet al., 2022) and achieve new state-of-the-artresults on Xsum (Narayan et al., 2018) for allROUGE scores (ROUGE-1 49.17, ROUGE-227.20, ROUGE-L 40.98).

McQueen: a Benchmark for Multimodal Conversational Query Rewrite
Yifei Yuan | Chen Shi | Runze Wang | Liyi Chen | Feijun Jiang | Yuan You | Wai Lam

The task of query rewrite aims to convert an in-context query to its fully-specified version where ellipsis and coreference are completed and referred-back according to the history context. Although much progress has been made, less efforts have been paid to real scenario conversations that involve drawing information from more than one modalities. In this paper, we propose the task of multimodal conversational query rewrite (McQR), which performs query rewrite under the multimodal visual conversation setting. We collect a large-scale dataset named McQueen based on manual annotation, which contains 15k visual conversations and over 80k queries where each one is associated with a fully-specified rewrite version. In addition, for entities appearing in the rewrite, we provide the corresponding image box annotation. We then use the McQueen dataset to benchmark a state-of-the-art method for effectively tackling the McQR task, which is based on a multimodal pre-trained model with pointer generator. Extensive experiments are performed to demonstrate the effectiveness of our model on this task.

Self-supervised Graph Masking Pre-training for Graph-to-Text Generation
Jiuzhou Han | Ehsan Shareghi

Large-scale pre-trained language models (PLMs) have advanced Graph-to-Text (G2T) generation by processing the linearised version of a graph. However, the linearisation is known to ignore the structural information. Additionally, PLMs are typically pre-trained on free text which introduces domain mismatch between pre-training and downstream G2T generation tasks. To address these shortcomings, we propose graph masking pre-training strategies that neither require supervision signals nor adjust the architecture of the underlying pre-trained encoder-decoder model. When used with a pre-trained T5, our approach achieves new state-of-the-art results on WebNLG+2020 and EventNarrative G2T generation datasets. Our method also shows to be very effective in the low-resource setting.

Improving Stability of Fine-Tuning Pretrained Language Models via Component-Wise Gradient Norm Clipping
Chenghao Yang | Xuezhe Ma

Fine-tuning over large pretrained language models (PLMs) has established many state-of-the-art results. Despite its superior performance, such fine-tuning can be unstable, resulting in significant variance in performance and potential risks for practical applications. Previous works have attributed such instability to the catastrophic forgetting problem in the top layers of PLMs, which indicates iteratively fine-tuning layers in a top-down manner is a promising solution. In this paper, we first point out that this method does not always work out due to the different convergence speeds of different layers/modules. Inspired by this observation, we propose a simple component-wise gradient norm clipping method to adjust the convergence speed for different components. Experiment results demonstrate that our method achieves consistent improvements in terms of generalization performance, convergence speed, and training stability. The codebase can be found at

Differentially Private Language Models for Secure Data Sharing
Justus Mattern | Zhijing Jin | Benjamin Weggenmann | Bernhard Schoelkopf | Mrinmaya Sachan

To protect the privacy of individuals whose data is being shared, it is of high importance to develop methods allowing researchers and companies to release textual data while providing formal privacy guarantees to its originators. In the field of NLP, substantial efforts have been directed at building mechanisms following the framework of local differential privacy, thereby anonymizing individual text samples before releasing them. In practice, these approaches are often dissatisfying in terms of the quality of their output language due to the strong noise required for local differential privacy. In this paper, we approach the problem at hand using global differential privacy, particularly by training a generative language model in a differentially private manner and consequently sampling data from it. Using natural language prompts and a new prompt-mismatch loss, we are able to create highly accurate and fluent textual datasets taking on specific desired attributes such as sentiment or topic and resembling statistical properties of the training data. We perform thorough experiments indicating that our synthetic datasets do not leak information from our original data and are of high language quality and highly suitable for training models for further analysis on real-world data. Notably, we also demonstrate that training classifiers on private synthetic data outperforms directly training classifiers with DP-SGD.

Conditional set generation using Seq2seq models
Aman Madaan | Dheeraj Rajagopal | Niket Tandon | Yiming Yang | Antoine Bosselut

Conditional set generation learns a mapping from an input sequence of tokens to a set. Several NLP tasks, such as entity typing and dialogue emotion tagging, are instances of set generation. Seq2Seq models are a popular choice to model set generation but they treat a set as a sequence and do not fully leverage its key properties, namely order-invariance and cardinality. We propose a novel algorithm for effectively sampling informative orders over the combinatorial space of label orders. Further, we jointly model the set cardinality and output by listing the set size as the first element and taking advantage of the autoregressive factorization used by Seq2Seq models. Our method is a model-independent data augmentation approach that endows any Seq2Seq model with the signals of order-invariance and cardinality. Training a Seq2Seq model on this new augmented data (without any additional annotations), gets an average relative improvement of 20% for four benchmarks datasets across models spanning from BART-base, T5-11B, and GPT-3. We will release all code and data upon acceptance.

Analyzing and Evaluating Faithfulness in Dialogue Summarization
Bin Wang | Chen Zhang | Yan Zhang | Yiming Chen | Haizhou Li

Dialogue summarization is abstractive in nature, making it suffer from factual errors. The factual correctness of summaries has the highest priority before practical applications. Many efforts have been made to improve faithfulness in text summarization. However, there is a lack of systematic study on dialogue summarization systems. In this work, we first perform the fine-grained human analysis on the faithfulness of dialogue summaries and observe that over 35% of generated summaries are faithfully inconsistent respective the source dialogues. Furthermore, we present a new model-level faithfulness evaluation method. It examines generation models with multi-choice questions created by rule-based transformations. Experimental results show that our evaluation schema is a strong proxy for the factual correctness of summarization models. The human-annotated faithfulness samples and the evaluation toolkit are released to facilitate future research toward faithful dialogue summarization.

Twist Decoding: Diverse Generators Guide Each Other
Jungo Kasai | Keisuke Sakaguchi | Ronan Le Bras | Hao Peng | Ximing Lu | Dragomir Radev | Yejin Choi | Noah A. Smith

Many language generation models are now available for a wide range of generation tasks, including machine translation and summarization. Combining such diverse models may lead to further progress, but ensembling generation models is challenging during inference: conventional ensembling methods (e.g., shallow fusion) require that the models share vocabulary/tokenization schemes. We introduce Twist decoding, a simple and general text generation algorithm that benefits from diverse models at inference time. Our method does not assume the vocabulary, tokenization or even generation order is shared. Our extensive evaluations on machine translation and scientific paper summarization demonstrate that Twist decoding substantially outperforms each model decoded in isolation over various scenarios, including cases where domain-specific and general-purpose models are both available. Twist decoding also consistently outperforms the popular reranking heuristic where output candidates from one model are rescored by another. We hope that our work will encourage researchers and practitioners to examine generation models collectively, not just independently, and to seek out models with complementary strengths to the currently available models.

Exploring Representation-level Augmentation for Code Search
Haochen Li | Chunyan Miao | Cyril Leung | Yanxian Huang | Yuan Huang | Hongyu Zhang | Yanlin Wang

Code search, which aims at retrieving the most relevant code fragment for a given natural language query, is a common activity in software development practice. Recently, contrastive learning is widely used in code search research, where many data augmentation approaches for source code (e.g., semantic-preserving program transformation) are proposed to learn better representations. However, these augmentations are at the raw-data level, which requires additional code analysis in the preprocessing stage and additional training cost in the training stage. In this paper, we explore augmentation methods that augment data (both code and query) at representation level which does not require additional data processing and training, and based on this we propose a general format of representation-level augmentation that unifies existing methods. Then, we propose three new augmentation methods (linear extrapolation, binary interpolation, and Gaussian scaling) based on the general format. Furthermore, we theoretically analyze the advantages of the proposed augmentation methods over traditional contrastive learning methods on code search. We experimentally evaluate the proposed representation-level augmentation methods with state-of-the-art code search models on a large-scale public dataset consisting of six programming languages. The experimental results show that our approach can consistently boost the performance of the studied code search models.

Learning Semantic Textual Similarity via Topic-informed Discrete Latent Variables
Erxin Yu | Lan Du | Yuan Jin | Zhepei Wei | Yi Chang

Recently, discrete latent variable models have received a surge of interest in both Natural Language Processing (NLP) and Computer Vision (CV), attributed to their comparable performance to the continuous counterparts in representation learning, while being more interpretable in their predictions. In this paper, we develop a topic-informed discrete latent variable model for semantic textual similarity, which learns a shared latent space for sentence-pair representation via vector quantization. Compared with previous models limited to local semantic contexts, our model can explore richer semantic information via topic modeling. We further boost the performance of semantic similarity by injecting the quantized representation into a transformer-based language model with a well-designed semantic-driven attention mechanism. We demonstrate, through extensive experiments across various English language datasets, that our model is able to surpass several strong neural baselines in semantic textual similarity tasks.

STRUDEL: Structured Dialogue Summarization for Dialogue Comprehension
Borui Wang | Chengcheng Feng | Arjun Nair | Madelyn Mao | Jai Desai | Asli Celikyilmaz | Haoran Li | Yashar Mehdad | Dragomir Radev

Abstractive dialogue summarization has long been viewed as an important standalone task in natural language processing, but no previous work has explored the possibility of whether abstractive dialogue summarization can also be used as a means to boost an NLP system’s performance on other important dialogue comprehension tasks. In this paper, we propose a novel type of dialogue summarization task - STRUctured DiaLoguE Summarization (STRUDEL) - that can help pre-trained language models to better understand dialogues and improve their performance on important dialogue comprehension tasks. In contrast to the holistic approach taken by the traditional free-form abstractive summarization task for dialogues, STRUDEL aims to decompose and imitate the hierarchical, systematic and structured mental process that we human beings usually go through when understanding and analyzing dialogues, and thus has the advantage of being more focused, specific and instructive for dialogue comprehension models to learn from. We further introduce a new STRUDEL dialogue comprehension modeling framework that integrates STRUDEL into a dialogue reasoning module over transformer encoder language models to improve their dialogue comprehension ability. In our empirical experiments on two important downstream dialogue comprehension tasks - dialogue question answering and dialogue response prediction - we demonstrate that our STRUDEL dialogue comprehension models can significantly improve the dialogue comprehension performance of transformer encoder language models.

Competency-Aware Neural Machine Translation: Can Machine Translation Know its Own Translation Quality?
Pei Zhang | Baosong Yang | Hao-Ran Wei | Dayiheng Liu | Kai Fan | Luo Si | Jun Xie

Neural machine translation (NMT) is often criticized for failures that happenwithout awareness. The lack of competency awareness makes NMT untrustworthy. This is in sharp contrast to human translators who give feedback or conduct further investigations whenever they are in doubt about predictions. To fill this gap, we propose a novel competency-aware NMT by extending conventional NMT with a self-estimator, offering abilities to translate a source sentence and estimate its competency.The self-estimator encodes the information of the decoding procedure and then examines whether it can reconstruct the original semantics of the source sentence. Experimental results on four translation tasks demonstrate that the proposed method not only carries out translation tasks intact but also delivers outstanding performance on quality estimation.Without depending on any reference or annotated data typically required by state-of-the-art metric and quality estimation methods, our model yields an even higher correlation with human quality judgments than a variety of aforementioned methods, such as BLEURT, COMET, and BERTScore. Quantitative and qualitative analyses show better robustness of competency awareness in our model.

PASTA: Table-Operations Aware Fact Verification via Sentence-Table Cloze Pre-training
Zihui Gu | Ju Fan | Nan Tang | Preslav Nakov | Xiaoman Zhao | Xiaoyong Du

Fact verification has attracted a lot of attention recently, e.g., in journalism, marketing, and policymaking, as misinformation and dis- information can sway one’s opinion and affect one’s actions. While fact-checking is a hard task in general, in many cases, false statements can be easily debunked based on analytics over tables with reliable information. Hence, table- based fact verification has recently emerged as an important and growing research area. Yet, progress has been limited due to the lack of datasets that can be used to pre-train language models (LMs) to be aware of common table operations, such as aggregating a column or comparing tuples. To bridge this gap, this paper introduces PASTA for table-based fact verification via pre-training with synthesized sentence–table cloze questions. In particular, we design six types of common sentence–table cloze tasks, including Filter, Aggregation, Superlative, Comparative, Ordinal, and Unique, based on which we synthesize a large corpus consisting of 1.2 million sentence–table pairs from WikiTables. PASTA uses a recent pre-trained LM, DeBERTaV3, and further pre- trains it on our corpus. Our experimental results show that PASTA achieves new state-of-the-art (SOTA) performance on two table-based fact verification datasets TabFact and SEM-TAB- FACTS. In particular, on the complex set of TabFact, which contains multiple operations, PASTA largely outperforms previous SOTA by 4.7% (85.6% vs. 80.9%), and the gap between PASTA and human performance on the small test set is narrowed to just 1.5% (90.6% vs. 92.1%).

Sentiment-Aware Word and Sentence Level Pre-training for Sentiment Analysis
Shuai Fan | Chen Lin | Haonan Li | Zhenghao Lin | Jinsong Su | Hang Zhang | Yeyun Gong | JIan Guo | Nan Duan

Most existing pre-trained language representation models (PLMs) are sub-optimal in sentiment analysis tasks, as they capture the sentiment information from word-level while under-considering sentence-level information. In this paper, we propose SentiWSP, a novel Sentiment-aware pre-trained language model with combined Word-level and Sentence-level Pre-training tasks.The word level pre-training task detects replaced sentiment words, via a generator-discriminator framework, to enhance the PLM’s knowledge about sentiment words.The sentence level pre-training task further strengthens the discriminator via a contrastive learning framework, with similar sentences as negative samples, to encode sentiments in a sentence.Extensive experimental results show that SentiWSP achieves new state-of-the-art performance on various sentence-level and aspect-level sentiment classification benchmarks. We have made our code and model publicly available at

Towards Multi-Modal Sarcasm Detection via Hierarchical Congruity Modeling with Knowledge Enhancement
Hui Liu | Wenya Wang | Haoliang Li

Sarcasm is a linguistic phenomenon indicating a discrepancy between literal meanings and implied intentions. Due to its sophisticated nature, it is usually difficult to be detected from the text itself. As a result, multi-modal sarcasm detection has received more and more attention in both academia and industries. However, most existing techniques only modeled the atomic-level inconsistencies between the text input and its accompanying image, ignoring more complex compositions for both modalities. Moreover, they neglected the rich information contained in external knowledge, e.g., image captions. In this paper, we propose a novel hierarchical framework for sarcasm detection by exploring both the atomic-level congruity based on multi-head cross attentions and the composition-level congruity based on graph neural networks, where a post with low congruity can be identified as sarcasm. In addition, we exploit the effect of various knowledge resources for sarcasm detection. Evaluation results on a public multi-modal sarcasm detection dataset based on Twitter demonstrate the superiority of our proposed model.

Efficiently Tuned Parameters Are Task Embeddings
Wangchunshu Zhou | Canwen Xu | Julian McAuley

Intermediate-task transfer can benefit a wide range of NLP tasks with properly selected source datasets. However, it is computationally infeasible to experiment with all intermediate transfer combinations, making choosing a useful source task a challenging problem. In this paper, we anticipate that task-specific parameters updated in parameter-efficient tuning methods are likely to encode task-specific information. Therefore, such parameters can be predictive for inter-task transferability. Thus, we propose to exploit these efficiently tuned parameters as off-the-shelf task embeddings for the efficient selection of source datasets for intermediate-task transfer. We experiment with 11 text classification tasks and 11 question answering tasks. Experimental results show that our approach consistently outperforms existing inter-task transferability prediction methods while being conceptually simple and computationally efficient. Our analysis also reveals that the ability of efficiently tuned parameters on transferability prediction is disentangled with their in-task performance. This allows us to use parameters from early checkpoints as task embeddings to further improve efficiency.

COPEN: Probing Conceptual Knowledge in Pre-trained Language Models
Hao Peng | Xiaozhi Wang | Shengding Hu | Hailong Jin | Lei Hou | Juanzi Li | Zhiyuan Liu | Qun Liu

Conceptual knowledge is fundamental to human cognition and knowledge bases. However, existing knowledge probing works only focus on evaluating factual knowledge of pre-trained language models (PLMs) and ignore conceptual knowledge. Since conceptual knowledge often appears as implicit commonsense behind texts, designing probes for conceptual knowledge is hard. Inspired by knowledge representation schemata, we comprehensively evaluate conceptual knowledge of PLMs by designing three tasks to probe whether PLMs organize entities by conceptual similarities, learn conceptual properties, and conceptualize entities in contexts, respectively. For the tasks, we collect and annotate 24k data instances covering 393 concepts, which is COPEN, a COnceptual knowledge Probing bENchmark. Extensive experiments on different sizes and types of PLMs show that existing PLMs systematically lack conceptual knowledge and suffer from various spurious correlations. We believe this is a critical bottleneck for realizing human-like cognition in PLMs. COPEN and our codes are publicly released at

Capturing Global Structural Information in Long Document Question Answering with Compressive Graph Selector Network
Yuxiang Nie | Heyan Huang | Wei Wei | Xian-Ling Mao

Long document question answering is a challenging task due to its demands for complex reasoning over long text. Previous works usually take long documents as non-structured flat texts or only consider the local structure in long documents. However, these methods usually ignore the global structure of the long document, which is essential for long-range understanding. To tackle this problem, we propose Compressive Graph Selector Network (CGSN) to capture the global structure in a compressive and iterative manner. The proposed model mainly focuses on the evidence selection phase of long document question answering. Specifically, it consists of three modules: local graph network, global graph network and evidence memory network. Firstly, the local graph network builds the graph structure of the chunked segment in token, sentence, paragraph and segment levels to capture the short-term dependency of the text. Secondly, the global graph network selectively receives the information of each level from the local graph, compresses them into the global graph nodes and applies graph attention to the global graph nodes to build the long-range reasoning over the entire text in an iterative way. Thirdly, the evidence memory network is designed to alleviate the redundancy problem in the evidence selection by saving the selected result in the previous steps. Extensive experiments show that the proposed model outperforms previous methods on two datasets.

Structural generalization is hard for sequence-to-sequence models
Yuekun Yao | Alexander Koller

Sequence-to-sequence (seq2seq) models have been successful across many NLP tasks,including ones that require predicting linguistic structure. However, recent work on compositional generalization has shown that seq2seq models achieve very low accuracy in generalizing to linguistic structures that were not seen in training. We present new evidence that this is a general limitation of seq2seq models that is present not just in semantic parsing, but also in syntactic parsing and in text-to-text tasks, and that this limitation can often be overcome by neurosymbolic models that have linguistic knowledge built in. We further report on some experiments that give initial answers on the reasons for these limitations.

Contrastive Learning enhanced Author-Style Headline Generation
Hui Liu | Weidong Guo | Yige Chen | Xiangyang Li

Headline generation is a task of generating an appropriate headline for a given article, which can be further used for machine-aided writing or enhancing the click-through ratio. Current works only use the article itself in the generation, but have not taken the writing style of headlines into consideration. In this paper, we propose a novel Seq2Seq model called CLH3G (Contrastive Learning enhanced Historical Headlines based Headline Generation) which can use the historical headlines of the articles that the author wrote in the past to improve the headline generation of current articles. By taking historical headlines into account, we can integrate the stylistic features of the author into our model, and generate a headline not only appropriate for the article, but also consistent with the author’s style. In order to efficiently learn the stylistic features of the author, we further introduce a contrastive learning based auxiliary task for the encoder of our model. Besides, we propose two methods to use the learned stylistic features to guide both the pointer and the decoder during the generation. Experimental results show that historical headlines of the same user can improve the headline generation significantly, and both the contrastive learning module and the two style features fusion methods can further boost the performance.

Multi-Granularity Optimization for Non-Autoregressive Translation
Yafu Li | Leyang Cui | Yongjing Yin | Yue Zhang

Despite low latency, non-autoregressive machine translation (NAT) suffers severe performance deterioration due to the naive independence assumption. This assumption is further strengthened by cross-entropy loss, which encourages a strict match between the hypothesis and the reference token by token. To alleviate this issue, we propose multi-granularity optimization for NAT, which collects model behaviours on translation segments of various granularities and integrates feedback for backpropagation. Experiments on four WMT benchmarks show that the proposed method significantly outperforms the baseline models trained with cross-entropy loss, and achieves the best performance on WMT’16 En⇔Ro and highly competitive results on WMT’14 En⇔De for fully non-autoregressive translation.

Super-NaturalInstructions: Generalization via Declarative Instructions on 1600+ NLP Tasks
Yizhong Wang | Swaroop Mishra | Pegah Alipoormolabashi | Yeganeh Kordi | Amirreza Mirzaei | Atharva Naik | Arjun Ashok | Arut Selvan Dhanasekaran | Anjana Arunkumar | David Stap | Eshaan Pathak | Giannis Karamanolakis | Haizhi Lai | Ishan Purohit | Ishani Mondal | Jacob Anderson | Kirby Kuznia | Krima Doshi | Kuntal Kumar Pal | Maitreya Patel | Mehrad Moradshahi | Mihir Parmar | Mirali Purohit | Neeraj Varshney | Phani Rohitha Kaza | Pulkit Verma | Ravsehaj Singh Puri | Rushang Karia | Savan Doshi | Shailaja Keyur Sampat | Siddhartha Mishra | Sujan Reddy A | Sumanta Patro | Tanay Dixit | Xudong Shen

How well can NLP models generalize to a variety of unseen tasks when provided with task instructions? To address this question, we first introduce Super-NaturalInstructions, a benchmark of 1,616 diverse NLP tasks and their expert-written instructions. Our collection covers 76 distinct task types, including but not limited to classification, extraction, infilling, sequence tagging, text rewriting, and text composition. This large and diverse collection of tasks enables rigorous benchmarking of cross-task generalization under instructions—training models to follow instructions on a subset of tasks and evaluating them on the remaining unseen ones.Furthermore, we build Tk-Instruct, a transformer model trained to follow a variety of in-context instructions (plain language task definitions or k-shot examples). Our experiments show that Tk-Instruct outperforms existing instruction-following models such as InstructGPT by over 9% on our benchmark despite being an order of magnitude smaller. We further analyze generalization as a function of various scaling parameters, such as the number of observed tasks, the number of instances per task, and model sizes. We hope our dataset and model facilitate future progress towards more general-purpose NLP models.

MetaFill: Text Infilling for Meta-Path Generation on Heterogeneous Information Networks
Zequn Liu | Kefei Duan | Junwei Yang | Hanwen Xu | Ming Zhang | Sheng Wang

Heterogeneous information network (HIN) is essential to study complicated networks containing multiple edge types and node types. Meta-path, a sequence of node types and edge types, is the core technique to embed HINs. Since manually curating meta-paths is time-consuming, there is a pressing need to develop automated meta-path generation approaches. Existing meta-path generation approaches cannot fully exploit the rich textual information in HINs, such as node names and edge type names. To address this problem, we propose MetaFill, a text-infilling-based approach for meta-path generation. The key idea of MetaFill is to formulate meta-path identification problem as a word sequence infilling problem, which can be advanced by pretrained language models (PLMs). We observed the superior performance of MetaFill against existing meta-path generation methods and graph embedding methods that do not leverage meta-paths in both link prediction and node classification on two real-world HIN datasets. We further demonstrated how MetaFill can accurately classify edges in the zero-shot setting, where existing approaches cannot generate any meta-paths. MetaFill exploits PLMs to generate meta-paths for graph embedding, opening up new avenues for language model applications in graph analysis.

DRLK: Dynamic Hierarchical Reasoning with Language Model and Knowledge Graph for Question Answering
Miao Zhang | Rufeng Dai | Ming Dong | Tingting He

In recent years, Graph Neural Network (GNN) approaches with enhanced knowledge graphs (KG) perform well in question answering (QA) tasks. One critical challenge is how to effectively utilize interactions between the QA context and KG. However, existing work only adopts the identical QA context representation to interact with multiple layers of KG, which results in a restricted interaction. In this paper, we propose DRLK (Dynamic Hierarchical Reasoning with Language Model and Knowledge Graphs), a novel model that utilizes dynamic hierarchical interactions between the QA context and KG for reasoning. DRLK extracts dynamic hierarchical features in the QA context, and performs inter-layer and intra-layer interactions on each iteration, allowing the KG representation to be grounded with the hierarchical features of the QA context. We conduct extensive experiments on four benchmark datasets in medical QA and commonsense reasoning. The experimental results demonstrate that DRLK achieves state-of-the-art performances on two benchmark datasets and performs competitively on the others.

AEG: Argumentative Essay Generation via A Dual-Decoder Model with Content Planning
Jianzhu Bao | Yasheng Wang | Yitong Li | Fei Mi | Ruifeng Xu

Argument generation is an important but challenging task in computational argumentation.Existing studies have mainly focused on generating individual short arguments, while research on generating long and coherent argumentative essays is still under-explored.In this paper, we propose a new task, Argumentative Essay Generation (AEG).Given a writing prompt, the goal of AEG is to automatically generate an argumentative essay with strong persuasiveness.We construct a large-scale dataset, ArgEssay, for this new task and establish a strong model based on a dual-decoder Transformer architecture.Our proposed model contains two decoders, a planning decoder (PD) and a writing decoder (WD), where PD is used to generate a sequence for essay content planning and WD incorporates the planning information to write an essay.Further, we pre-train this model on a large news dataset to enhance the plan-and-write paradigm.Automatic and human evaluation results show that our model can generate more coherent and persuasive essays with higher diversity and less repetition compared to several baselines.

BotsTalk: Machine-sourced Framework for Automatic Curation of Large-scale Multi-skill Dialogue Datasets
Minju Kim | Chaehyeong Kim | Yong Ho Song | Seung-won Hwang | Jinyoung Yeo

To build open-domain chatbots that are able to use diverse communicative skills, we propose a novel framework BotsTalk, where multiple agents grounded to the specific target skills participate in a conversation to automatically annotate multi-skill dialogues. We further present Blended Skill BotsTalk (BSBT), a large-scale multi-skill dialogue dataset comprising 300K conversations. Through extensive experiments, we demonstrate that our dataset can be effective for multi-skill dialogue systems which require an understanding of skill blending as well as skill grounding. Our code and data are available at

Wider & Closer: Mixture of Short-channel Distillers for Zero-shot Cross-lingual Named Entity Recognition
Jun-Yu Ma | Beiduo Chen | Jia-Chen Gu | Zhenhua Ling | Wu Guo | Quan Liu | Zhigang Chen | Cong Liu

Zero-shot cross-lingual named entity recognition (NER) aims at transferring knowledge from annotated and rich-resource data in source languages to unlabeled and lean-resource data in target languages. Existing mainstream methods based on the teacher-student distillation framework ignore the rich and complementary information lying in the intermediate layers of pre-trained language models, and domain-invariant information is easily lost during transfer. In this study, a mixture of short-channel distillers (MSD) method is proposed to fully interact the rich hierarchical information in the teacher model and to transfer knowledge to the student model sufficiently and efficiently. Concretely, a multi-channel distillation framework is designed for sufficient information transfer by aggregating multiple distillers as a mixture. Besides, an unsupervised method adopting parallel domain adaptation is proposed to shorten the channels between the teacher and student models to preserve domain-invariant features. Experiments on four datasets across nine languages demonstrate that the proposed method achieves new state-of-the-art performance on zero-shot cross-lingual NER and shows great generalization and compatibility across languages and fields.

An Efficient Memory-Augmented Transformer for Knowledge-Intensive NLP Tasks
Yuxiang Wu | Yu Zhao | Baotian Hu | Pasquale Minervini | Pontus Stenetorp | Sebastian Riedel

Access to external knowledge is essential for many natural language processing tasks, such as question answering and dialogue. Existing methods often rely on a parametric model that stores knowledge in its parameters, or use a retrieval-augmented model that has access to an external knowledge source. Parametric and retrieval-augmented models have complementary strengths in terms of computational efficiency and predictive accuracy. To combine the strength of both approaches, we propose the Efficient Memory-Augmented Transformer (EMAT) – it encodes external knowledge into a key-value memory and exploits the fast maximum inner product search for memory querying. We also introduce pre-training tasks that allow EMAT to encode informative key-value representations, and to learn an implicit strategy to integrate multiple memory slots into the transformer. Experiments on various knowledge-intensive tasks such as question answering and dialogue datasets show that, simply augmenting parametric models (T5-base) using our method produces more accurate results (e.g., 25.8 → 44.3 EM on NQ) while retaining a high throughput (e.g., 1000 queries/s on NQ). Compared to retrieval-augmented models, EMAT runs substantially faster across the board and produces more accurate results on WoW and ELI5.

Supervised Prototypical Contrastive Learning for Emotion Recognition in Conversation
Xiaohui Song | Longtao Huang | Hui Xue | Songlin Hu

Capturing emotions within a conversation plays an essential role in modern dialogue systems. However, the weak correlation between emotions and semantics brings many challenges to emotion recognition in conversation (ERC). Even semantically similar utterances, the emotion may vary drastically depending on contexts or speakers. In this paper, we propose a Supervised Prototypical Contrastive Learning (SPCL) loss for the ERC task. Leveraging the Prototypical Network, the SPCL targets at solving the imbalanced classification problem through contrastive learning and does not require a large batch size. Meanwhile, we design a difficulty measure function based on the distance between classes and introduce curriculum learning to alleviate the impact of extreme samples. We achieve state-of-the-art results on three widely used benchmarks. Further, we conduct analytical experiments to demonstrate the effectiveness of our proposed SPCL and curriculum learning strategy.

RuCoLA: Russian Corpus of Linguistic Acceptability
Vladislav Mikhailov | Tatiana Shamardina | Max Ryabinin | Alena Pestova | Ivan Smurov | Ekaterina Artemova

Linguistic acceptability (LA) attracts the attention of the research community due to its many uses, such as testing the grammatical knowledge of language models and filtering implausible texts with acceptability classifiers.However, the application scope of LA in languages other than English is limited due to the lack of high-quality resources.To this end, we introduce the Russian Corpus of Linguistic Acceptability (RuCoLA), built from the ground up under the well-established binary LA approach. RuCoLA consists of 9.8k in-domain sentences from linguistic publications and 3.6k out-of-domain sentences produced by generative models. The out-of-domain set is created to facilitate the practical use of acceptability for improving language generation.Our paper describes the data collection protocol and presents a fine-grained analysis of acceptability classification experiments with a range of baseline approaches.In particular, we demonstrate that the most widely used language models still fall behind humans by a large margin, especially when detecting morphological and semantic errors. We release RuCoLA, the code of experiments, and a public leaderboard to assess the linguistic competence of language models for Russian.

Complex Hyperbolic Knowledge Graph Embeddings with Fast Fourier Transform
Huiru Xiao | Xin Liu | Yangqiu Song | Ginny Wong | Simon See

The choice of geometric space for knowledge graph (KG) embeddings can have significant effects on the performance of KG completion tasks. The hyperbolic geometry has been shown to capture the hierarchical patterns due to its tree-like metrics, which addressed the limitations of the Euclidean embedding models. Recent explorations of the complex hyperbolic geometry further improved the hyperbolic embeddings for capturing a variety of hierarchical structures. However, the performance of the hyperbolic KG embedding models for non-transitive relations is still unpromising, while the complex hyperbolic embeddings do not deal with multi-relations. This paper aims to utilize the representation capacity of the complex hyperbolic geometry in multi-relational KG embeddings. To apply the geometric transformations which account for different relations and the attention mechanism in the complex hyperbolic space, we propose to use the fast Fourier transform (FFT) as the conversion between the real and complex hyperbolic space. Constructing the attention-based transformations in the complex space is very challenging, while the proposed Fourier transform-based complex hyperbolic approaches provide a simple and effective solution. Experimental results show that our methods outperform the baselines, including the Euclidean and the real hyperbolic embedding models.

Towards Knowledge-Intensive Text-to-SQL Semantic Parsing with Formulaic Knowledge
Longxu Dou | Yan Gao | Xuqi Liu | Mingyang Pan | Dingzirui Wang | Wanxiang Che | Dechen Zhan | Min-Yen Kan | Jian-Guang Lou

In this paper, we study the problem of knowledge-intensive text-to-SQL, in which domain knowledge is necessary to parse expert questions into SQL queries over domain-specific tables. We formalize this scenario by building a new benchmark KnowSQL consisting of domain-specific questions covering various domains. We then address this problem by representing formulaic knowledge rather than by annotating additional data examples. More concretely, we construct a formulaic knowledge bank as a domain knowledge base and propose a framework (ReGrouP) to leverage this formulaic knowledge during parsing. Experiments using ReGrouP demonstrate a significant 28.2% improvement overall on KnowSQL.

Should We Ban English NLP for a Year?
Anders Søgaard

Around two thirds of NLP research at top venues is devoted exclusively to developing technology for speakers of English, most speech data comes from young urban speakers, and most texts used to train language models come from male writers. These biases feed into consumer technologies to widen existing inequality gaps, not only within, but also across, societies. Many have argued that it is almost impossible to mitigate inequality amplification. I argue that, on the contrary, it is quite simple to do so, and that counter-measures would have little-to-no negative impact, except for, perhaps, in the very short term.

LittleBird: Efficient Faster & Longer Transformer for Question Answering
Minchul Lee | Kijong Han | Myeong Cheol Shin

BERT has shown a lot of sucess in a wide variety of NLP tasks. But it has a limitation dealing with long inputs due to its attention mechanism. Longformer, ETC and BigBird addressed this issue and effectively solved the quadratic dependency problem.However we find that these models are not sufficient, and propose LittleBird, a novel model based on BigBird with improved speed and memory footprint while maintaining accuracy.In particular, we devise a more flexible and efficient position representation method based on Attention with Linear Biases(ALiBi). We also show that replacing the method of global information represented in the BigBird with pack and unpack attention is more effective.The proposed model can work on long inputs even after being pre-trained on short inputs, and can be trained efficiently reusing existing pre-trained language model for short inputs. This is a significant benefit for low-resource languages where large amounts of long text data are difficult to obtain.As a result, our experiments show that LittleBird works very well in a variety of languages, achieving high performance in question answering tasks, particularly in KorQuAD2.0, Korean Question Answering Dataset for long paragraphs.

WeTS: A Benchmark for Translation Suggestion
Zhen Yang | Fandong Meng | Yingxue Zhang | Ernan Li | Jie Zhou

Translation suggestion (TS), which provides alternatives for specific words or phrases given the entire documents generated by machine translation (MT), has been proven to play a significant role in post-editing (PE). There are two main pitfalls for existing researches in this line. First, most conventional works only focus on the overall performance of PE but ignore the exact performance of TS, which makes the progress of PE sluggish and less explainable; Second, as no publicly available golden dataset exists to support in-depth research for TS, almost all of the previous works conduct experiments on their in-house datasets or the noisy datasets built automatically, which makes their experiments hard to be reproduced and compared. To break these limitations mentioned above and spur the research in TS, we create a benchmark dataset, called WeTS, which is a golden corpus annotated by expert translators on four translation directions. Apart from the golden corpus, we also propose several methods to generate synthetic corpora which can be used to improve the performance substantially through pre-training. As for the model, we propose the segment-aware self-attention based Transformer for TS. Experimental results show that our approach achieves the best results on all four directions, including English-to-German, German-to-English, Chinese-to-English, and English-to-Chinese.

Discrete Cross-Modal Alignment Enables Zero-Shot Speech Translation
Chen Wang | Yuchen Liu | Boxing Chen | Jiajun Zhang | Wei Luo | Zhongqiang Huang | Chengqing Zong

End-to-end Speech Translation (ST) aims at translating the source language speech into target language text without generating the intermediate transcriptions. However, the training of end-to-end methods relies on parallel ST data, which are difficult and expensive to obtain. Fortunately, the supervised data for automatic speech recognition (ASR) and machine translation (MT) are usually more accessible, making zero-shot speech translation a potential direction. Existing zero-shot methods fail to align the two modalities of speech and text into a shared semantic space, resulting in much worse performance compared to the supervised ST methods. In order to enable zero-shot ST, we propose a novel Discrete Cross-Modal Alignment (DCMA) method that employs a shared discrete vocabulary space to accommodate and match both modalities of speech and text. Specifically, we introduce a vector quantization module to discretize the continuous representations of speech and text into a finite set of virtual tokens, and use ASR data to map corresponding speech and text to the same virtual token in a shared codebook. This way, source language speech can be embedded in the same semantic space as the source language text, which can be then transformed into target language text with an MT module. Experiments on multiple language pairs demonstrate that our zero-shot ST method significantly improves the SOTA, and even performers on par with the strong supervised ST baselines.

Abstractive Summarization Guided by Latent Hierarchical Document Structure
Yifu Qiu | Shay B. Cohen

Sequential abstractive neural summarizers often do not use the underlying structure in the input article or dependencies between the input sentences. This structure is essential to integrate and consolidate information from different parts of the text. To address this shortcoming, we propose a hierarchy-aware graph neural network (HierGNN) which captures such dependencies through three main steps: 1) learning a hierarchical document structure through a latent structure tree learned by a sparse matrix-tree computation; 2) propagating sentence information over this structure using a novel message-passing node propagation mechanism to identify salient information; 3) using graph-level attention to concentrate the decoder on salient information. Experiments confirm HierGNN improves strong sequence models such as BART, with a 0.55 and 0.75 margin in average ROUGE-1/2/L for CNN/DM and XSum. Further human evaluation demonstrates that summaries produced by our model are more relevant and less redundant than the baselines, into which HierGNN is incorporated. We also find HierGNN synthesizes summaries by fusing multiple source sentences more, rather than compressing a single source sentence, and that it processes long inputs more effectively.

Explainable Question Answering based on Semantic Graph by Global Differentiable Learning and Dynamic Adaptive Reasoning
Jianguo Mao | Wenbin Jiang | Xiangdong Wang | Hong Liu | Yu Xia | Yajuan Lyu | QiaoQiao She

Multi-hop Question Answering is an agent task for testing the reasoning ability. With the development of pre-trained models, the implicit reasoning ability has been surprisingly improved and can even surpass human performance. However, the nature of the black box hinders the construction of explainable intelligent systems. Several researchers have explored explainable neural-symbolic reasoning methods based on question decomposition techniques. The undifferentiable symbolic operations and the error propagation in the reasoning process lead to poor performance. To alleviate it, we propose a simple yet effective Global Differentiable Learning strategy to explore optimal reasoning paths from the latent probability space so that the model learns to solve intermediate reasoning processes without expert annotations. We further design a Dynamic Adaptive Reasoner to enhance the generalization of unseen questions. Our method achieves 17% improvements in F1-score against BreakRC and shows better interpretability. We take a step forward in building interpretable reasoning methods.

DuReader-Retrieval: A Large-scale Chinese Benchmark for Passage Retrieval from Web Search Engine
Yifu Qiu | Hongyu Li | Yingqi Qu | Ying Chen | QiaoQiao She | Jing Liu | Hua Wu | Haifeng Wang

In this paper, we present DuReader-retrieval, a large-scale Chinese dataset for passage retrieval. DuReader-retrieval contains more than 90K queries and over 8M unique passages from a commercial search engine. To alleviate the shortcomings of other datasets and ensure the quality of our benchmark, we (1) reduce the false negatives in development and test sets by manually annotating results pooled from multiple retrievers, and (2) remove the training queries that are semantically similar to the development and testing queries. Additionally, we provide two out-of-domain testing sets for cross-domain evaluation, as well as a set of human translated queries for for cross-lingual retrieval evaluation. The experiments demonstrate that DuReader-retrieval is challenging and a number of problems remain unsolved, such as the salient phrase mismatch and the syntactic mismatch between queries and paragraphs. These experiments also show that dense retrievers do not generalize well across domains, and cross-lingual retrieval is essentially challenging. DuReader-retrieval is publicly available at

Pair-Based Joint Encoding with Relational Graph Convolutional Networks for Emotion-Cause Pair Extraction
Junlong Liu | Xichen Shang | Qianli Ma

Emotion-cause pair extraction (ECPE) aims to extract emotion clauses and corresponding cause clauses, which have recently received growing attention. Previous methods sequentially encode features with a specified order. They first encode the emotion and cause features for clause extraction and then combine them for pair extraction. This lead to an imbalance in inter-task feature interaction where features extracted later have no direct contact with the former. To address this issue, we propose a novel **P**air-**B**ased **J**oint **E**ncoding (**PBJE**) network, which generates pairs and clauses features simultaneously in a joint feature encoding manner to model the causal relationship in clauses. PBJE can balance the information flow among emotion clauses, cause clauses and pairs. From a multi-relational perspective, we construct a heterogeneous undirected graph and apply the Relational Graph Convolutional Network (RGCN) to capture the multiplex relationship between clauses and the relationship between pairs and clauses. Experimental results show that PBJE achieves state-of-the-art performance on the Chinese benchmark corpus.

Affective Knowledge Enhanced Multiple-Graph Fusion Networks for Aspect-based Sentiment Analysis
Siyu Tang | Heyan Chai | Ziyi Yao | Ye Ding | Cuiyun Gao | Binxing Fang | Qing Liao

Aspect-based sentiment analysis aims to identify sentiment polarity of social media users toward different aspects. Most recent methods adopt the aspect-centric latent tree to connect aspects and their corresponding opinion words, thinking that would facilitate establishing the relationship between aspects and opinion words.However, these methods ignore the roles of syntax dependency relation labels and affective semantic information in determining the sentiment polarity, resulting in the wrong prediction.In this paper, we propose a novel multi-graph fusion network (MGFN) based on latent graph to leverage the richer syntax dependency relation label information and affective semantic information of words.Specifically, we construct a novel syntax-aware latent graph (SaLG) to fully leverage the syntax dependency relation label information to facilitate the learning of sentiment representations. Subsequently, a multi-graph fusion module is proposed to fuse semantic information of surrounding contexts of aspects adaptively. Furthermore, we design an affective refinement strategy to guide the MGFN to capture significant affective clues. Extensive experiments on three datasets demonstrate that our MGFN model outperforms all state-of-the-art methods and verify the effectiveness of our model.

IndicNLG Benchmark: Multilingual Datasets for Diverse NLG Tasks in Indic Languages
Aman Kumar | Himani Shrotriya | Prachi Sahu | Amogh Mishra | Raj Dabre | Ratish Puduppully | Anoop Kunchukuttan | Mitesh M. Khapra | Pratyush Kumar

Natural Language Generation (NLG) for non-English languages is hampered by the scarcity of datasets in these languages. We present the IndicNLG Benchmark, a collection of datasets for benchmarking NLG for 11 Indic languages. We focus on five diverse tasks, namely, biography generation using Wikipedia infoboxes, news headline generation, sentence summarization, paraphrase generation and, question generation. We describe the created datasets and use them to benchmark the performance of several monolingual and multilingual baselines that leverage pre-trained sequence-to-sequence models. Our results exhibit the strong performance of multilingual language-specific pre-trained models, and the utility of models trained on our dataset for other related NLG tasks. Our dataset creation methods can be easily applied to modest-resource languages as they involve simple steps such as scraping news articles and Wikipedia infoboxes, light cleaning, and pivoting through machine translation data. To the best of our knowledge, the IndicNLG Benchmark is the first NLG benchmark for Indic languages and the most diverse multilingual NLG dataset, with approximately 8M examples across 5 tasks and 11 languages. The datasets and models will be publicly available.

Improving Machine Translation with Phrase Pair Injection and Corpus Filtering
Akshay Batheja | Pushpak Bhattacharyya

In this paper, we show that the combination of Phrase Pair Injection and Corpus Filtering boosts the performance of Neural Machine Translation (NMT) systems. We extract parallel phrases and sentences from the pseudo-parallel corpus and augment it with the parallel corpus to train the NMT models. With the proposed approach, we observe an improvement in the Machine Translation (MT) system for 3 low-resource language pairs, Hindi-Marathi, English-Marathi, and English-Pashto, and 6 translation directions by up to 2.7 BLEU points, on the FLORES test data. These BLEU score improvements are over the models trained using the whole pseudo-parallel corpus augmented with the parallel corpus.

An Anchor-based Relative Position Embedding Method for Cross-Modal Tasks
Ya Wang | Xingwu Sun | Lian Fengzong | ZhanHui Kang | Chengzhong Xu Xu

Position Embedding (PE) is essential for transformer to capture the sequence ordering of input tokens. Despite its general effectiveness verified in Natural Language Processing (NLP) and Computer Vision (CV), its application in cross-modal tasks remains unexplored and suffers from two challenges: 1) the input text tokens and image patches are not aligned, 2) the encoding space of each modality is different, making it unavailable for feature comparison. In this paper, we propose a unified position embedding method for these problems, called AnChor-basEd Relative Position Embedding (ACE-RPE), in which we first introduce an anchor locating mechanism to bridge the semantic gap and locate anchors from different modalities. Then we conduct the distance calculation of each text token and image patch by computing their shortest paths from the located anchors. Last, we embed the anchor-based distance to guide the computation of cross-attention. In this way, it calculates cross-modal relative position embedding for cross-modal transformer. Benefiting from ACE-RPE, our method obtains new SOTA results on a wide range of benchmarks, such as Image-Text Retrieval on MS-COCO and Flickr30K, Visual Entailment on SNLI-VE, Visual Reasoning on NLVR2 and Weakly-supervised Visual Grounding on RefCOCO+.

Norm-based Noisy Corpora Filtering and Refurbishing in Neural Machine Translation
Yu Lu | Jiajun Zhang

Recent advances in neural machine translation depend on massive parallel corpora, which are collected from any open source without much guarantee of quality. It stresses the need for noisy corpora filtering, but existing methods are insufficient to solve this issue. They spend much time ensembling multiple scorers trained on clean bitexts, unavailable for low-resource languages in practice. In this paper, we propose a norm-based noisy corpora filtering and refurbishing method with no external data and costly scorers. The noisy and clean samples are separated based on how much information from the source and target sides the model requires to fit the given translation. For the unparallel sentence, the target-side history translation is much more important than the source context, contrary to the parallel ones. The amount of these two information flows can be measured by norms of source-/target-side context vectors. Moreover, we propose to reuse the discovered noisy data by generating pseudo labels via online knowledge distillation. Extensive experiments show that our proposed filtering method performs comparably with state-of-the-art noisy corpora filtering techniques but is more efficient and easier to operate. Noisy sample refurbishing further enhances the performance by making the most of the given data.

TeleMelody: Lyric-to-Melody Generation with a Template-Based Two-Stage Method
Zeqian Ju | Peiling Lu | Xu Tan | Rui Wang | Chen Zhang | Songruoyao Wu | Kejun Zhang | Xiang-Yang Li | Tao Qin | Tie-Yan Liu

Lyric-to-melody generation is an important task in automatic songwriting. Previous lyric-to-melody generation systems usually adopt end-to-end models that directly generate melodies from lyrics, which suffer from several issues: 1) lack of paired lyric-melody training data; 2) lack of control on generated melodies. In this paper, we develop TeleMelody, a two-stage lyric-to-melody generation system with music template (e.g., tonality, chord progression, rhythm pattern, and cadence) to bridge the gap between lyrics and melodies (i.e., the system consists of a lyric-to-template module and a template-to-melody module). TeleMelody has two advantages. First, it is data efficient. The template-to-melody module is trained in a self-supervised way (i.e., the source template is extracted from the target melody) that does not need any lyric-melody paired data. The lyric-to-template module is made up of some rules and a lyric-to-rhythm model, which is trained with paired lyric-rhythm data that is easier to obtain than paired lyric-melody data. Second, it is controllable. The design of the template ensures that the generated melodies can be controlled by adjusting the musical elements in the template. Both subjective and objective experimental evaluations demonstrate that TeleMelody generates melodies with higher quality, better controllability, and less requirement on paired lyric-melody data than previous generation systems.

SEEN: Structured Event Enhancement Network for Explainable Need Detection of Information Recall Assistance
You-En Lin | An-Zi Yen | Hen-Hsen Huang | Hsin-Hsi Chen

When recalling life experiences, people often forget or confuse life events, which necessitates information recall services. Previous work on information recall focuses on providing such assistance reactively, i.e., by retrieving the life event of a given query. Proactively detecting the need for information recall services is rarely discussed. In this paper, we use a human-annotated life experience retelling dataset to detect the right time to trigger the information recall service. We propose a pilot model—structured event enhancement network (SEEN) that detects life event inconsistency, additional information in life events, and forgotten events. A fusing mechanism is also proposed to incorporate event graphs of stories and enhance the textual representations. To explain the need detection results, SEEN simultaneously provides support evidence by selecting the related nodes from the event graph. Experimental results show that SEEN achieves promising performance in detecting information needs. In addition, the extracted evidence can be served as complementary information to remind users what events they may want to recall.

Rethinking Style Transformer with Energy-based Interpretation: Adversarial Unsupervised Style Transfer using a Pretrained Model
Hojun Cho | Dohee Kim | Seungwoo Ryu | ChaeHun Park | Hyungjong Noh | Jeong-in Hwang | Minseok Choi | Edward Choi | Jaegul Choo

Style control, content preservation, and fluency determine the quality of text style transfer models. To train on a nonparallel corpus, several existing approaches aim to deceive the style discriminator with an adversarial loss. However, adversarial training significantly degrades fluency compared to the other two metrics. In this work, we explain this phenomenon using energy-based interpretation, and leverage a pretrained language model to improve fluency. Specifically, we propose a novel approach which applies the pretrained language model to the text style transfer framework by restructuring the discriminator and the model itself, allowing the generator and the discriminator to also take advantage of the power of the pretrained model. We evaluated our model on three public benchmarks GYAFC, Amazon, and Yelp and achieved state-of-the-art performance on the overall metrics.

Towards Robust k-Nearest-Neighbor Machine Translation
Hui Jiang | Ziyao Lu | Fandong Meng | Chulun Zhou | Jie Zhou | Degen Huang | Jinsong Su

k-Nearest-Neighbor Machine Translation (kNN-MT) becomes an important research direction of NMT in recent years. Its main idea is to retrieve useful key-value pairs from an additional datastore to modify translations without updating the NMT model. However, the underlying retrieved noisy pairs will dramatically deteriorate the model performance. In this paper, we conduct a preliminary study and find that this problem results from not fully exploiting the prediction of the NMT model. To alleviate the impact of noise, we propose a confidence-enhanced kNN-MT model with robust training. Concretely, we introduce the NMT confidence to refine the modeling of two important components of kNN-MT: kNN distribution and the interpolation weight. Meanwhile we inject two types of perturbations into the retrieved pairs for robust training. Experimental results on four benchmark datasets demonstrate that our model not only achieves significant improvements over current kNN-MT models, but also exhibits better robustness. Our code is available at

Tiny-NewsRec: Effective and Efficient PLM-based News Recommendation
Yang Yu | Fangzhao Wu | Chuhan Wu | Jingwei Yi | Qi Liu

News recommendation is a widely adopted technique to provide personalized news feeds for the user. Recently, pre-trained language models (PLMs) have demonstrated the great capability of natural language understanding and benefited news recommendation via improving news modeling. However, most existing works simply finetune the PLM with the news recommendation task, which may suffer from the known domain shift problem between the pre-training corpus and downstream news texts. Moreover, PLMs usually contain a large volume of parameters and have high computational overhead, which imposes a great burden on low-latency online services. In this paper, we propose Tiny-NewsRec, which can improve both the effectiveness and the efficiency of PLM-based news recommendation. We first design a self-supervised domain-specific post-training method to better adapt the general PLM to the news domain with a contrastive matching task between news titles and news bodies. We further propose a two-stage knowledge distillation method to improve the efficiency of the large PLM-based news recommendation model while maintaining its performance. Multiple teacher models originated from different time steps of our post-training procedure are used to transfer comprehensive knowledge to the student model in both its post-training stage and finetuning stage. Extensive experiments on two real-world datasets validate the effectiveness and efficiency of our method.

TABS: Efficient Textual Adversarial Attack for Pre-trained NL Code Model Using Semantic Beam Search
YunSeok Choi | Hyojun Kim | Jee-Hyong Lee

As pre-trained models have shown successful performance in program language processing as well as natural language processing, adversarial attacks on these models also attract attention.However, previous works on black-box adversarial attacks generated adversarial examples in a very inefficient way with simple greedy search. They also failed to find out better adversarial examples because it was hard to reduce the search space without performance loss.In this paper, we propose TABS, an efficient beam search black-box adversarial attack method. We adopt beam search to find out better adversarial examples, and contextual semantic filtering to effectively reduce the search space. Contextual semantic filtering reduces the number of candidate adversarial words considering the surrounding context and the semantic similarity.Our proposed method shows good performance in terms of attack success rate, the number of queries, and semantic similarity in attacking models for two tasks: NL code search classification and retrieval tasks.

Investigating the Robustness of Natural Language Generation from Logical Forms via Counterfactual Samples
Chengyuan Liu | Leilei Gan | Kun Kuang | Fei Wu

The aim of Logic2Text is to generate controllable and faithful texts conditioned on tables and logical forms, which not only requires a deep understanding of the tables and logical forms, but also warrants symbolic reasoning over the tables according to the logical forms. State-of-the-art methods based on pre-trained models have achieved remarkable performance on the standard test dataset. However, we question whether these methods really learn how to perform logical reasoning, rather than just relying on the spurious correlations between the headers of the tables and operators of the logical form. To verify this hypothesis, we manually construct a set of counterfactual samples, which modify the original logical forms to generate counterfactual logical forms with rare co-occurred headers and operators and corresponding counterfactual references. SOTA methods give much worse results on these counterfactual samples compared with the results on the original test dataset, which verifies our hypothesis. To deal with this problem, we firstly analyze this bias from a causal perspective, based on which we propose two approaches to reduce the model’s reliance on the shortcut. The first one incorporates the hierarchical structure of the logical forms into the model. The second one exploits automatically generated counterfactual data for training. Automatic and manual experimental results on the original test dataset and counterfactual dataset show that our method is effective to alleviate the spurious correlation. Our work points out the weakness of current methods and takes a further step toward developing Logic2Text models with real logical reasoning ability.

Helping the Weak Makes You Strong: Simple Multi-Task Learning Improves Non-Autoregressive Translators
Xinyou Wang | Zaixiang Zheng | Shujian Huang

Recently, non-autoregressive (NAR) neural machine translation models have received increasing attention due to their efficient parallel decoding.However, the probabilistic framework of NAR models necessitates conditional independence assumption on target sequences, falling short of characterizing human language data.This drawback results in less informative learning signals for NAR models under conventional MLE training, thereby yielding unsatisfactory accuracy compared to their autoregressive (AR) counterparts.In this paper, we propose a simple and model-agnostic multi-task learning framework to provide more informative learning signals.During training stage, we introduce a set of sufficiently weak AR decoders that solely rely on the information provided by NAR decoder to make prediction, forcing the NAR decoder to become stronger or else it will be unable to support its weak AR partners.Experiments on WMT and IWSLT datasets show that our approach can consistently improve accuracy of multiple NAR baselines without adding any additional decoding overhead.

RACE: Retrieval-augmented Commit Message Generation
Ensheng Shi | Yanlin Wang | Wei Tao | Lun Du | Hongyu Zhang | Shi Han | Dongmei Zhang | Hongbin Sun

Commit messages are important for software development and maintenance. Many neural network-based approaches have been proposed and shown promising results on automatic commit message generation. However, the generated commit messages could be repetitive or redundant. In this paper, we propose RACE, a new retrieval-augmented neural commit message generation method, which treats the retrieved similar commit as an exemplar and leverages it to generate an accurate commit message. As the retrieved commit message may not always accurately describe the content/intent of the current code diff, we also propose an exemplar guider, which learns the semantic similarity between the retrieved and current code diff and then guides the generation of commit message based on the similarity. We conduct extensive experiments on a large public dataset with five programming languages. Experimental results show that RACE can outperform all baselines. Furthermore, RACE can boost the performance of existing Seq2Seq models in commit message generation.

PLOG: Table-to-Logic Pretraining for Logical Table-to-Text Generation
Ao Liu | Haoyu Dong | Naoaki Okazaki | Shi Han | Dongmei Zhang

Logical table-to-text generation is a task that involves generating logically faithful sentences from tables, which requires models to derive logical-level facts from table records via logical inference. It raises a new challenge on the logical-level content planning of table-to-text models. However, directly learning the logical inference knowledge from table-text pairs is very difficult for neural models because of the ambiguity of natural language and the scarcity of parallel data. Hence even large-scale pre-trained language models present low logical fidelity on logical table-to-text. In this work, we propose a Pretrained Logical Form Generator (PLOG) framework to improve generation fidelity. Specifically, PLOG is first pretrained on a table-to-logical-form generation (table-to-logic) task, then finetuned on downstream table-to-text tasks. The logical forms are formally defined with unambiguous semantics. Hence we can collect a large amount of accurate logical forms from tables without human annotation. In addition, PLOG can learn logical inference from table-logic pairs much more reliably than from table-text pairs. To evaluate our model, we further collect a controlled logical table-to-text dataset CONTLOG based on an existing dataset. On two benchmarks, LOGICNLG and CONTLOG, PLOG outperforms strong baselines by a large margin on the logical fidelity, demonstrating the effectiveness of table-to-logic pretraining.

GHAN: Graph-Based Hierarchical Aggregation Network for Text-Video Retrieval
Yahan Yu | Bojie Hu | Yu Li

Text-video retrieval focuses on two aspects: cross-modality interaction and video-language encoding. Currently, the mainstream approach is to train a joint embedding space for multimodal interactions. However, there are structural and semantic differences between text and video, making this approach challenging for fine-grained understanding. In order to solve this, we propose an end-to-end graph-based hierarchical aggregation network for text-video retrieval according to the hierarchy possessed by text and video. We design a token-level weighted network to refine intra-modality representations and construct a graph-based message passing attention network for global-local alignment across modality. We conduct experiments on the public datasets MSR-VTT-9K, MSR-VTT-7K and MSVD, and achieve Recall@1 of 73.0%, 65.6%, and 64.0% , which is 25.7%, 16.5%, and 14.2% better than the current state-of-the-art model.

MuRAG: Multimodal Retrieval-Augmented Generator for Open Question Answering over Images and Text
Wenhu Chen | Hexiang Hu | Xi Chen | Pat Verga | William Cohen

While language Models store a massive amount of world knowledge implicitly in their parameters, even very large models often fail to encode information about rare entities and events, while incurring huge computational costs. Recently, retrieval-augmented models, such as REALM, RAG, and RETRO, have incorporated world knowledge into language generation by leveraging an external non-parametric index and have demonstrated impressive performance with constrained model sizes. However, these methods are restricted to retrieving only textual knowledge, neglecting the ubiquitous amount of knowledge in other modalities like images – much of which contains information not covered by any text. To address this limitation, we propose the first Multimodal Retrieval-Augmented Transformer (MuRAG), which accesses an external non-parametric multimodal memory to augment language generation. MuRAG is pre-trained with a mixture of large-scale image-text and text-only corpora using a joint contrastive and generative loss. We perform experiments on two different datasets that require retrieving and reasoning over both images and text to answer a given query: WebQA, and MultimodalQA. Our results show that MuRAG achieves state-of-the-art accuracy, outperforming existing models by 10-20% absolute on both datasets and under both distractor and full-wiki settings.

PHEE: A Dataset for Pharmacovigilance Event Extraction from Text
Zhaoyue Sun | Jiazheng Li | Gabriele Pergola | Byron Wallace | Bino John | Nigel Greene | Joseph Kim | Yulan He

The primary goal of drug safety researchers and regulators is to promptly identify adverse drug reactions. Doing so may in turn prevent or reduce the harm to patients and ultimately improve public health. Evaluating and monitoring drug safety (i.e., pharmacovigilance) involves analyzing an ever growing collection of spontaneous reports from health professionals, physicians, and pharmacists, and information voluntarily submitted by patients. In this scenario, facilitating analysis of such reports via automation has the potential to rapidly identify safety signals. Unfortunately, public resources for developing natural language models for this task are scant. We present PHEE, a novel dataset for pharmacovigilance comprising over 5000 annotated events from medical case reports and biomedical literature, making it the largest such public dataset to date. We describe the hierarchical event schema designed to provide coarse and fine-grained information about patients’ demographics, treatments and (side) effects. Along with the discussion of the dataset, we present a thorough experimental evaluation of current state-of-the-art approaches for biomedical event extraction, point out their limitations, and highlight open challenges to foster future research in this area.

OTSeq2Set: An Optimal Transport Enhanced Sequence-to-Set Model for Extreme Multi-label Text Classification
Jie Cao | Yin Zhang

Extreme multi-label text classification (XMTC) is the task of finding the most relevant subset labels from an extremely large-scale label collection. Recently, some deep learning models have achieved state-of-the-art results in XMTC tasks. These models commonly predict scores for all labels by a fully connected layer as the last layer of the model. However, such models can’t predict a relatively complete and variable-length label subset for each document, because they select positive labels relevant to the document by a fixed threshold or take top k labels in descending order of scores. A less popular type of deep learning models called sequence-to-sequence (Seq2Seq) focus on predicting variable-length positive labels in sequence style. However, the labels in XMTC tasks are essentially an unordered set rather than an ordered sequence, the default order of labels restrains Seq2Seq models in training. To address this limitation in Seq2Seq, we propose an autoregressive sequence-to-set model for XMTC tasks named OTSeq2Set. Our model generates predictions in student-forcing scheme and is trained by a loss function based on bipartite matching which enables permutation-invariance. Meanwhile, we use the optimal transport distance as a measurement to force the model to focus on the closest labels in semantic label space. Experiments show that OTSeq2Set outperforms other competitive baselines on 4 benchmark datasets. Especially, on the Wikipedia dataset with 31k labels, it outperforms the state-of-the-art Seq2Seq method by 16.34% in micro-F1 score. The code is available at

SimQA: Detecting Simultaneous MT Errors through Word-by-Word Question Answering
HyoJung Han | Marine Carpuat | Jordan Boyd-Graber

Detractors of neural machine translation admit that while its translations are fluent, it sometimes gets key facts wrong. This is particularly important in simultaneous interpretation where translations have to be provided as fast as possible: before a sentence is complete. Yet, evaluations of simultaneous machine translation (SimulMT) fail to capture if systems correctly translate the most salient elements of a question: people, places, and dates. To address this problem, we introduce a downstream word-by-word question answering evaluation task (SimQA): given a source language question, translate the question word by word into the target language, and answer as soon as possible. SimQA jointly measures whether the SimulMT models translate the question quickly and accurately, and can reveal shortcomings in existing neural systems—hallucinating or omitting facts.

Discovering Low-rank Subspaces for Language-agnostic Multilingual Representations
Zhihui Xie | Handong Zhao | Tong Yu | Shuai Li

Large pretrained multilingual language models (ML-LMs) have shown remarkable capabilities of zero-shot cross-lingual transfer, without direct cross-lingual supervision. While these results are promising, follow-up works found that, within the multilingual embedding spaces, there exists strong language identity information which hinders the expression of linguistic factors shared across languages. For semantic tasks like cross-lingual sentence retrieval, it is desired to remove such language identity signals to fully leverage semantic information. In this work, we provide a novel view of projecting away language-specific factors from a multilingual embedding space. Specifically, we discover that there exists a low-rank subspace that primarily encodes information irrelevant to semantics (e.g., syntactic information). To identify this subspace, we present a simple but effective unsupervised method based on singular value decomposition with multiple monolingual corpora as input. Once the subspace is found, we can directly project the original embeddings into the null space to boost language agnosticism without finetuning. We systematically evaluate our method on various tasks including the challenging language-agnostic QA retrieval task. Empirical results show that applying our method consistently leads to improvements over commonly used ML-LMs.

Rethinking the Authorship Verification Experimental Setups
Florin Brad | Andrei Manolache | Elena Burceanu | Antonio Barbalau | Radu Tudor Ionescu | Marius Popescu

One of the main drivers of the recent advances in authorship verification is the PAN large-scale authorship dataset. Despite generating significant progress in the field, inconsistent performance differences between the closed and open test sets have been reported. To this end, we improve the experimental setup by proposing five new public splits over the PAN dataset, specifically designed to isolate and identify biases related to the text topic and to the author’s writing style. We evaluate several BERT-like baselines on these splits, showing that such models are competitive with authorship verification state-of-the-art methods. Furthermore, using explainable AI, we find that these baselines are biased towards named entities. We show that models trained without the named entities obtain better results and generalize better when tested on DarkReddit, our new dataset for authorship verification.

Borrowing Human Senses: Comment-Aware Self-Training for Social Media Multimodal Classification
Chunpu Xu | Jing Li

Social media is daily creating massive multimedia content with paired image and text, presenting the pressing need to automate the vision and language understanding for various multimodal classification tasks. Compared to the commonly researched visual-lingual data, social media posts tend to exhibit more implicit image-text relations. To better glue the cross-modal semantics therein, we capture hinting features from user comments, which are retrieved via jointly leveraging visual and lingual similarity. Afterwards, the classification tasks are explored via self-training in a teacher-student framework, motivated by the usually limited labeled data scales in existing benchmarks. Substantial experiments are conducted on four multimodal social media benchmarks for image-text relation classification, sarcasm detection, sentiment classification, and hate speech detection. The results show that our method further advances the performance of previous state-of-the-art models, which do not employ comment modeling or self-training.

Training Language Models with Memory Augmentation
Zexuan Zhong | Tao Lei | Danqi Chen

Recent work has improved language models (LMs) remarkably by equipping them with a non-parametric memory component. However, most existing approaches only introduce mem-ories at testing time or represent them using a separately trained encoder, resulting in suboptimal training of the language model. In this work, we present TRIME, a novel yet simple training approach designed for training LMs with memory augmentation. Our approach uses a training objective that directly takes in-batch examples as accessible memory. We also present new methods for memory construction and data batching, which are used for adapting to different sets of memories—local, long-term, and external memory—at testing time. We evaluate TRIME on multiple language modeling and machine translation benchmarks and show that it is able to achieve significant improvements across all the settings. Concretely, TRIME reduces the perplexity from 18.70 to 15.37 on WIKITEXT-103, by effectively leveraging a large memory set from the training corpus. Compared to standard LM training, TRIME adds negligible computational overhead and is compatible with different neural architectures, making it a versatile solution for training memory-augmented LMs.

Data-Efficient Strategies for Expanding Hate Speech Detection into Under-Resourced Languages
Paul Röttger | Debora Nozza | Federico Bianchi | Dirk Hovy

Hate speech is a global phenomenon, but most hate speech datasets so far focus on English-language content. This hinders the development of more effective hate speech detection models in hundreds of languages spoken by billions across the world. More data is needed, but annotating hateful content is expensive, time-consuming and potentially harmful to annotators. To mitigate these issues, we explore data-efficient strategies for expanding hate speech detection into under-resourced languages. In a series of experiments with mono- and multilingual models across five non-English languages, we find that 1) a small amount of target-language fine-tuning data is needed to achieve strong performance, 2) the benefits of using more such data decrease exponentially, and 3) initial fine-tuning on readily-available English data can partially substitute target-language data and improve model generalisability. Based on these findings, we formulate actionable recommendations for hate speech detection in low-resource language settings.

Dimension Reduction for Efficient Dense Retrieval via Conditional Autoencoder
Zhenghao Liu | Han Zhang | Chenyan Xiong | Zhiyuan Liu | Yu Gu | Xiaohua Li

Dense retrievers encode queries and documents and map them in an embedding space using pre-trained language models. These embeddings need to be high-dimensional to fit training signals and guarantee the retrieval effectiveness of dense retrievers. However, these high-dimensional embeddings lead to larger index storage and higher retrieval latency. To reduce the embedding dimensions of dense retrieval, this paper proposes a Conditional Autoencoder (ConAE) to compress the high-dimensional embeddings to maintain the same embedding distribution and better recover the ranking features. Our experiments show that ConAE is effective in compressing embeddings by achieving comparable ranking performance with its teacher model and making the retrieval system more efficient. Our further analyses show that ConAE can alleviate the redundancy of the embeddings of dense retrieval with only one linear layer. All codes of this work are available at

Controlled Text Reduction
Aviv Slobodkin | Paul Roit | Eran Hirsch | Ori Ernst | Ido Dagan

Producing a reduced version of a source text, as in generic or focused summarization, inherently involves two distinct subtasks: deciding on targeted content and generating a coherent text conveying it. While some popular approaches address summarization as a single end-to-end task, prominent works support decomposed modeling for individual subtasks. Further, semi-automated text reduction is also very appealing, where users may identify targeted content while models would generate a corresponding coherent summary.In this paper, we focus on the second subtask, of generating coherent text given pre-selected content. Concretely, we formalize Controlled Text Reduction as a standalone task, whose input is a source text with marked spans of targeted content (“highlighting”).A model then needs to generate a coherent text that includes all and only the target information.We advocate the potential of such models, both for modular fully-automatic summarization, as well as for semi-automated human-in-the-loop use cases.Facilitating proper research, we crowdsource high-quality dev and test datasets for the task. Further, we automatically generate a larger “silver” training dataset from available summarization benchmarks, leveraging a pretrained summary-source alignment model.Finally, employing these datasets, we present a supervised baseline model, showing promising results and insightful analyses.

Questioning the Validity of Summarization Datasets and Improving Their Factual Consistency
Yanzhu Guo | Chloé Clavel | Moussa Kamal Eddine | Michalis Vazirgiannis

The topic of summarization evaluation has recently attracted a surge of attention due to the rapid development of abstractive summarization systems. However, the formulation of the task is rather ambiguous, neither the linguistic nor the natural language processing communities have succeeded in giving a mutually agreed-upon definition. Due to this lack of well-defined formulation, a large number of popular abstractive summarization datasets are constructed in a manner that neither guarantees validity nor meets one of the most essential criteria of summarization: factual consistency. In this paper, we address this issue by combining state-of-the-art factual consistency models to identify the problematic instances present in popular summarization datasets. We release SummFC, a filtered summarization dataset with improved factual consistency, and demonstrate that models trained on this dataset achieve improved performance in nearly all quality aspects. We argue that our dataset should become a valid benchmark for developing and evaluating summarization systems.

Invariant Language Modeling
Maxime Peyrard | Sarvjeet Ghotra | Martin Josifoski | Vidhan Agarwal | Barun Patra | Dean Carignan | Emre Kiciman | Saurabh Tiwary | Robert West

Modern pretrained language models are critical components of NLP pipelines. Yet, they suffer from spurious correlations, poor out-of-domain generalization, and biases.Inspired by recent progress in causal machine learning, in particular the invariant risk minimization (IRM) paradigm, we propose invariant language modeling, a framework for learning invariant representations that generalize better across multiple environments. In particular, we adapt a game-theoretic implementation of IRM (IRM-games) to language models, where the invariance emerges from a specific training schedule in which all the environments compete to optimize their own environment-specific loss by updating subsets of the model in a round-robin fashion.We focused on controlled experiments to precisely demonstrate the ability of our method to (i) remove structured noise, (ii) ignore specific spurious correlations without affecting global performance, and (iii) achieve better out-of-domain generalization.These benefits come with a negligible computational overhead compared to standard training, do not require changing the local loss, and can be applied to any language model. We believe this framework is promising to help mitigate spurious correlations and biases in language models.

AdaMix: Mixture-of-Adaptations for Parameter-efficient Model Tuning
Yaqing Wang | Sahaj Agarwal | Subhabrata Mukherjee | Xiaodong Liu | Jing Gao | Ahmed Hassan Awadallah | Jianfeng Gao

Standard fine-tuning of large pre-trained language models (PLMs) for downstream tasks requires updating hundreds of millions to billions of parameters, and storing a large copy of the PLM weights for every task resulting in increased cost for storing, sharing and serving the models. To address this, parameter-efficient fine-tuning (PEFT) techniques were introduced where small trainable components are injected in the PLM and updated during fine-tuning. We propose AdaMix as a general PEFT method that tunes a mixture of adaptation modules – given the underlying PEFT method of choice – introduced in each Transformer layer while keeping most of the PLM weights frozen. For instance, AdaMix can leverage a mixture of adapters like Houlsby or a mixture of low rank decomposition matrices like LoRA to improve downstream task performance over the corresponding PEFT methods for fully supervised and few-shot NLU and NLG tasks. Further, we design AdaMix such that it matches the same computational cost and the number of tunable parameters as the underlying PEFT method. By only tuning 0.1-0.2% of PLM parameters, we show that AdaMix outperforms SOTA parameter-efficient fine-tuning and full model fine-tuning for both NLU and NLG tasks.

How “Multi” is Multi-Document Summarization?
Ruben Wolhandler | Arie Cattan | Ori Ernst | Ido Dagan

The task of multi-document summarization (MDS) aims at models that, given multiple documents as input, are able to generate a summary that combines disperse information, originally spread __across__ these documents. Accordingly, it is expected that both reference summaries in MDS datasets, as well as system summaries, would indeed be based on such dispersed information. In this paper, we argue for quantifying and assessing this expectation. To that end, we propose an automated measure for evaluating the degree to which a summary is “disperse”, in the sense of the number of source documents needed to cover its content. We apply our measure to empirically analyze several popular MDS datasets, with respect to their reference summaries, as well as the output of state-of-the-art systems. Our results show that certain MDS datasets barely require combining information from multiple documents, where a single document often covers the full summary content. Overall, we advocate using our metric for assessing and improving the degree to which summarization datasets require combining multi-document information, and similarly how summarization models actually meet this challenge.

BioReader: a Retrieval-Enhanced Text-to-Text Transformer for Biomedical Literature
Giacomo Frisoni | Miki Mizutani | Gianluca Moro | Lorenzo Valgimigli

The latest batch of research has equipped language models with the ability to attend over relevant and factual information from non-parametric external sources, drawing a complementary path to architectural scaling. Besides mastering language, exploiting and contextualizing the latent world knowledge is crucial in complex domains like biomedicine. However, most works in the field rely on general-purpose models supported by databases like Wikipedia and Books. We introduce BioReader, the first retrieval-enhanced text-to-text model for biomedical natural language processing. Our domain-specific T5-based solution augments the input prompt by fetching and assembling relevant scientific literature chunks from a neural database with ≈60 million tokens centered on PubMed. We fine-tune and evaluate BioReader on a broad array of downstream tasks, significantly outperforming several state-of-the-art methods despite using up to 3x fewer parameters. In tandem with extensive ablation studies, we show that domain knowledge can be easily altered or supplemented to make the model generate correct predictions bypassing the retraining step and thus addressing the literature overload issue.

T-Modules: Translation Modules for Zero-Shot Cross-Modal Machine Translation
Paul-Ambroise Duquenne | Hongyu Gong | Benoît Sagot | Holger Schwenk

We present a new approach to perform zero-shot cross-modal transfer between speech and text for translation tasks. Multilingual speech and text are encoded in a joint fixed-size representation space. Then, we compare different approaches to decode these multimodal and multilingual fixed-size representations, enabling zero-shot translation between languages and modalities. All our models are trained without the need of cross-modal labeled translation data.Despite a fixed-size representation, we achieve very competitive results on several text and speech translation tasks. In particular, we significantly improve the state-of-the-art for zero-shot speech translation on Must-C. Incorporating a speech decoder in our framework, we introduce the first results for zero-shot direct speech-to-speech and text-to-speech translation.

LILA: A Unified Benchmark for Mathematical Reasoning
Swaroop Mishra | Matthew Finlayson | Pan Lu | Leonard Tang | Sean Welleck | Chitta Baral | Tanmay Rajpurohit | Oyvind Tafjord | Ashish Sabharwal | Peter Clark | Ashwin Kalyan

Mathematical reasoning skills are essential for general-purpose intelligentsystems to perform tasks from grocery shopping to climate modeling.Towards evaluating and improving AI systems in this domain, we proposeLILA, a unified mathematical reasoning benchmark consisting of 23 diversetasks along four dimensions:(i) mathematical abilities e.g., arithmetic, calculus (ii) language format e.g., question-answering, fill-in-the-blanks (iii) language diversity e.g., no language, simple language (iv) external knowledge e.g., commonsense, physics. We construct our benchmark by extending 20 datasets benchmark by collecting task instructions and solutions in the form of Python programs,thereby obtaining explainable solutions in addition to the correct answer.We additionally introduce two evaluation datasets to measure out-of-distribution performance and robustness to language perturbation.Finally, we introduce BHASKARA,a general-purpose mathematical reasoning model trained on LILA. Importantly, we find that multi-tasking leads to significant improvements (average relative improvement of 21.83% F1 score vs. single-task models),while the best performing model only obtains 60.40%,indicating the room for improvement in general mathematical reasoning and understanding.

Leveraging Affirmative Interpretations from Negation Improves Natural Language Understanding
Md Mosharaf Hossain | Eduardo Blanco

Negation poses a challenge in many natural language understanding tasks. Inspired by the fact that understanding a negated statement often requires humans to infer affirmative interpretations, in this paper we show that doing so benefits models for three natural language understanding tasks. We present an automated procedure to collect pairs of sentences with negation and their affirmative interpretations, resulting in over 150,000 pairs. Experimental results show that leveraging these pairs helps (a) T5 generate affirmative interpretations from negations in a previous benchmark, and (b) a RoBERTa-based classifier solve the task of natural language inference. We also leverage our pairs to build a plug-and-play neural generator that given a negated statement generates an affirmative interpretation. Then, we incorporate the pretrained generator into a RoBERTa-based classifier for sentiment analysis and show that doing so improves the results. Crucially, our proposal does not require any manual effort.

GraphQ IR: Unifying the Semantic Parsing of Graph Query Languages with One Intermediate Representation
Lunyiu Nie | Shulin Cao | Jiaxin Shi | Jiuding Sun | Qi Tian | Lei Hou | Juanzi Li | Jidong Zhai

Subject to the huge semantic gap between natural and formal languages, neural semantic parsing is typically bottlenecked by its complexity of dealing with both input semantics and output syntax. Recent works have proposed several forms of supplementary supervision but none is generalized across multiple formal languages. This paper proposes a unified intermediate representation for graph query languages, named GraphQ IR. It has a natural-language-like expression that bridges the semantic gap and formally defined syntax that maintains the graph structure. Therefore, a neural semantic parser can more precisely convert user queries into GraphQ IR, which can be later losslessly compiled into various downstream graph query languages. Extensive experiments on several benchmarks including KQA Pro, Overnight, GrailQA, and MetaQA-Cypher under the standard i.i.d., out-of-distribution, and low-resource settings validate GraphQ IR’s superiority over the previous state-of-the-arts with a maximum 11% accuracy improvement.

InforMask: Unsupervised Informative Masking for Language Model Pretraining
Nafis Sadeq | Canwen Xu | Julian McAuley

Masked language modeling is widely used for pretraining large language models for natural language understanding (NLU). However, random masking is suboptimal, allocating an equal masking rate for all tokens. In this paper, we propose InforMask, a new unsupervised masking strategy for training masked language models. InforMask exploits Pointwise Mutual Information (PMI) to select the most informative tokens to mask. We further propose two optimizations for InforMask to improve its efficiency. With a one-off preprocessing step, InforMask outperforms random masking and previously proposed masking strategies on the factual recall benchmark LAMA and the question answering benchmark SQuAD v1 and v2.

CTRLsum: Towards Generic Controllable Text Summarization
Junxian He | Wojciech Kryscinski | Bryan McCann | Nazneen Rajani | Caiming Xiong

Current summarization systems yield generic summaries that are disconnected from users’ preferences and expectations. To address this limitation, we present CTRLsum, a generic framework to control generated summaries through a set of keywords. During training keywords are extracted automatically without requiring additional human annotations. At test time CTRLsum features a control function to map control signal to keywords; through engineering the control function, the same trained model is able to be applied to control summaries on various dimensions, while neither affecting the model training process nor the pretrained models. We additionally explore the combination of keywords and text prompts for more control tasks. Experiments demonstrate the effectiveness of CTRLsum on three domains of summarization datasets and five control tasks: (1) entity-centric and (2) length-controllable summarization, (3) contribution summarization on scientific papers, (4) invention purpose summarization on patent filings, and (5) question-guided summarization on news articles. Moreover, when used in a standard, unconstrained summarization setting, CTRLsum is comparable or better than strong pretrained systems.

Missing Counter-Evidence Renders NLP Fact-Checking Unrealistic for Misinformation
Max Glockner | Yufang Hou | Iryna Gurevych

Misinformation emerges in times of uncertainty when credible information is limited. This is challenging for NLP-based fact-checking as it relies on counter-evidence, which may not yet be available. Despite increasing interest in automatic fact-checking, it is still unclear if automated approaches can realistically refute harmful real-world misinformation. Here, we contrast and compare NLP fact-checking with how professional fact-checkers combat misinformation in the absence of counter-evidence. In our analysis, we show that, by design, existing NLP task definitions for fact-checking cannot refute misinformation as professional fact-checkers do for the majority of claims. We then define two requirements that the evidence in datasets must fulfill for realistic fact-checking: It must be (1) sufficient to refute the claim and (2) not leaked from existing fact-checking articles. We survey existing fact-checking datasets and find that all of them fail to satisfy both criteria. Finally, we perform experiments to demonstrate that models trained on a large-scale fact-checking dataset rely on leaked evidence, which makes them unsuitable in real-world scenarios. Taken together, we show that current NLP fact-checking cannot realistically combat real-world misinformation because it depends on unrealistic assumptions about counter-evidence in the data.

A Framework for Adapting Pre-Trained Language Models to Knowledge Graph Completion
Justin Lovelace | Carolyn Rosé

Recent work has demonstrated that entity representations can be extracted from pre-trained language models to develop knowledge graph completion models that are more robust to the naturally occurring sparsity found in knowledge graphs. In this work, we conduct a comprehensive exploration of how to best extract and incorporate those embeddings into knowledge graph completion models. We explore the suitability of the extracted embeddings for direct use in entity ranking and introduce both unsupervised and supervised processing methods that can lead to improved downstream performance. We then introduce supervised embedding extraction methods that can extract more informative representations. We then synthesize our findings and develop a knowledge graph completion model that significantly outperforms recent neural models.

Mutual Information Alleviates Hallucinations in Abstractive Summarization
Liam van der Poel | Ryan Cotterell | Clara Meister

Despite significant progress in the quality of language generated from abstractive summarization models, these models still exhibit the tendency to hallucinate, i.e., output content not supported by the source document. A number of works have tried to fix—or at least uncover the source of—the problem with limited success. In this paper, we identify a simple criterion under which models are significantly more likely to assign more probability to hallucinated content during generation: high model uncertainty. This finding offers a potential explanation for hallucinations: models default to favoring text with high marginal probability, i.e., high-frequency occurrences in the training set, when uncertain about a continuation. It also motivates possible routes for real-time intervention during decoding to prevent such hallucinations. We propose a decoding strategy that switches to optimizing for pointwise mutual information of the source and target token—rather than purely the probability of the target token—when the model exhibits uncertainty. Experiments on the dataset show that our method decreases the probability of hallucinated tokens while maintaining the Rouge and BERT-S scores of top-performing decoding strategies.

Toward the Limitation of Code-Switching in Cross-Lingual Transfer
Yukun Feng | Feng Li | Philipp Koehn

Multilingual pretrained models have shown strong cross-lingual transfer ability. Some works used code-switching sentences, which consist of tokens from multiple languages, to enhance the cross-lingual representation further, and have shown success in many zero-shot cross-lingual tasks. However, code-switched tokens are likely to cause grammatical incoherence in newly substituted sentences, and negatively affect the performance on token-sensitive tasks, such as Part-of-Speech (POS) tagging and Named-Entity-Recognition (NER). This paper mitigates the limitation of the code-switching method by not only making the token replacement but considering the similarity between the context and the switched tokens so that the newly substituted sentences are grammatically consistent during both training and inference. We conduct experiments on cross-lingual POS and NER over 30+ languages, and demonstrate the effectiveness of our method by outperforming the mBERT by 0.95 and original code-switching method by 1.67 on F1 scores.

Syntactically Rich Discriminative Training: An Effective Method for Open Information Extraction
Frank Mtumbuka | Thomas Lukasiewicz

Open information extraction (OIE) is the task of extracting facts "(Subject, Relation, Object)” from natural language text. We propose several new methods for training neural OIE models in this paper. First, we propose a novel method for computing syntactically rich text embeddings using the structure of dependency trees. Second, we propose a new discriminative training approach to OIE in which tokens in the generated fact are classified as “real” or “fake”, i.e., those tokens that are in both the generated and gold tuples, and those that are only in the generated tuple but not in the gold tuple. We also address the issue of repetitive tokens in generated facts and improve the models’ ability to generate implicit facts. Our approach reduces repetitive tokens by a factor of 23%. Finally, we present paraphrased versions of the CaRB, OIE2016, and LSOIE datasets, and show that the models’ performance substantially improves when trained on augmented datasets. Our best model beats the SOTA of IMoJIE on the recent CaRB dataset, with an improvement of 39.63% in F1 score.

Transformer-based Entity Typing in Knowledge Graphs
Zhiwei Hu | Victor Gutierrez-Basulto | Zhiliang Xiang | Ru Li | Jeff Pan

We investigate the knowledge graph entity typing task which aims at inferring plausible entity types. In this paper, we propose a novel Transformer-based Entity Typing (TET) approach, effectively encoding the content of neighbours of an entity by means of a transformer mechanism. More precisely, TET is composed of three different mechanisms: a local transformer allowing to infer missing entity types by independently encoding the information provided by each of its neighbours; a global transformer aggregating the information of all neighbours of an entity into a single long sequence to reason about more complex entity types; and a context transformer integrating neighbours content in a differentiated way through information exchange between neighbour pairs, while preserving the graph structure. Furthermore, TET uses information about class membership of types to semantically strengthen the representation of an entity. Experiments on two real-world datasets demonstrate the superior performance of TET compared to the state-of-the-art.

NewsClaims: A New Benchmark for Claim Detection from News with Attribute Knowledge
Revanth Gangi Reddy | Sai Chetan Chinthakindi | Zhenhailong Wang | Yi Fung | Kathryn Conger | Ahmed ELsayed | Martha Palmer | Preslav Nakov | Eduard Hovy | Kevin Small | Heng Ji

Claim detection and verification are crucial for news understanding and have emerged as promising technologies for mitigating misinformation and disinformation in the news. However, most existing work has focused on claim sentence analysis while overlooking additional crucial attributes (e.g., the claimer and the main object associated with the claim).In this work, we present NewsClaims, a new benchmark for attribute-aware claim detection in the news domain. We extend the claim detection problem to include extraction of additional attributes related to each claim and release 889 claims annotated over 143 news articles. NewsClaims aims to benchmark claim detection systems in emerging scenarios, comprising unseen topics with little or no training data. To this end, we see that zero-shot and prompt-based baselines show promising performance on this benchmark, while still considerably behind human performance.

IsoVec: Controlling the Relative Isomorphism of Word Embedding Spaces
Kelly Marchisio | Neha Verma | Kevin Duh | Philipp Koehn

The ability to extract high-quality translation dictionaries from monolingual word embedding spaces depends critically on the geometric similarity of the spaces—their degree of “isomorphism.” We address the root-cause of faulty cross-lingual mapping: that word embedding training resulted in the underlying spaces being non-isomorphic. We incorporate global measures of isomorphism directly into the skipgram loss function, successfully increasing the relative isomorphism of trained word embedding spaces and improving their ability to be mapped to a shared cross-lingual space. The result is improved bilingual lexicon induction in general data conditions, under domain mismatch, and with training algorithm dissimilarities. We release IsoVec at

Adversarial Concept Erasure in Kernel Space
Shauli Ravfogel | Francisco Vargas | Yoav Goldberg | Ryan Cotterell

The representation space of neural models for textual data emerges in an unsupervised manner during training. Understanding how human-interpretable concepts, such as gender, are encoded in these representations would improve the ability of users to control the content of these representations and analyze the working of the models that rely on them. One prominent approach to the control problem is the identification and removal of linear concept subspaces – subspaces in the representation space that correspond to a given concept. While those are tractable and interpretable, neural network do not necessarily represent concepts in linear subspaces. We propose a kernelization of the recently-proposed linear concept-removal objective, and show that it is effective in guarding against the ability of certain nonlinear adversaries to recover the concept. Interestingly, our findings suggest that the division between linear and nonlinear models is overly simplistic: when considering the concept of binary gender and its neutralization, we do not find a single kernel space that exclusively contains all the concept-related information. It is therefore challenging to protect against all nonlinear adversaries at once.

The Authenticity Gap in Human Evaluation
Kawin Ethayarajh | Dan Jurafsky

Human ratings are the gold standard in NLG evaluation. The standard protocol is to collect ratings of generated text, average across annotators, and rank NLG systems by their average scores. However, little consideration has been given as to whether this approach faithfully captures human preferences. Analyzing this standard protocol through the lens of utility theory in economics, we identify the implicit assumptions it makes about annotators. These assumptions are often violated in practice, in which case annotator ratings cease to reflect their preferences. The most egregious violations come from using Likert scales, which provably reverse the direction of the true preference in certain cases. We suggest improvements to the standard protocol to make it more theoretically sound, but even in its improved form, it cannot be used to evaluate open-ended tasks like story generation. For the latter, we propose a new human evaluation protocol called system-level probabilistic assessment (SPA). When human evaluation of stories is done with SPA, we can recover the ordering of GPT-3 models by size, with statistically significant results. However, when human evaluation is done with the standard protocol, less than half of the expected preferences can be recovered (e.g., there is no significant difference between curie and davinci, despite using a highly powered test).

BERT in Plutarch’s Shadows
Ivan Yamshchikov | Alexey Tikhonov | Yorgos Pantis | Charlotte Schubert | Jürgen Jost

The extensive surviving corpus of the ancient scholar Plutarch of Chaeronea (ca. 45-120 CE) also contains several texts which, according to current scholarly opinion, did not originate with him and are therefore attributed to an anonymous author Pseudo-Plutarch. These include, in particular, the work Placita Philosophorum (Quotations and Opinions of the Ancient Philosophers), which is extremely important for the history of ancient philosophy. Little is known about the identity of that anonymous author and its relation to other authors from the same period. This paper presents a BERT language model for Ancient Greek. The model discovers previously unknown statistical properties relevant to these literary, philosophical, and historical problems and can shed new light on this authorship question. In particular, the Placita Philosophorum, together with one of the other Pseudo-Plutarch texts, shows similarities with the texts written by authors from an Alexandrian context (2nd/3rd century CE).

Leveraging Locality in Abstractive Text Summarization
Yixin Liu | Ansong Ni | Linyong Nan | Budhaditya Deb | Chenguang Zhu | Ahmed Hassan Awadallah | Dragomir Radev

Neural attention models have achieved significant improvements on many natural language processing tasks. However, the quadratic memory complexity of the self-attention module with respect to the input length hinders their applications in long text summarization. Instead of designing more efficient attention modules, we approach this problem by investigating if models with a restricted context can have competitive performance compared with the memory-efficient attention models that maintain a global context by treating the input as a single sequence. Our model is applied to individual pages, which contain parts of inputs grouped by the principle of locality, during both the encoding and decoding stages. We empirically investigated three kinds of locality in text summarization at different levels of granularity, ranging from sentences to documents. Our experimental results show that our model has a better performance compared with strong baseline models with efficient attention modules, and our analysis provides further insights into our locality-aware modeling strategy.

Salience Allocation as Guidance for Abstractive Summarization
Fei Wang | Kaiqiang Song | Hongming Zhang | Lifeng Jin | Sangwoo Cho | Wenlin Yao | Xiaoyang Wang | Muhao Chen | Dong Yu

Abstractive summarization models typically learn to capture the salient information from scratch implicitly.Recent literature adds extractive summaries as guidance for abstractive summarization models to provide hints of salient content and achieves better performance.However, extractive summaries as guidance could be over strict, leading to information loss or noisy signals.Furthermore, it cannot easily adapt to documents with various abstractiveness.As the number and allocation of salience content pieces varies, it is hard to find a fixed threshold deciding which content should be included in the guidance.In this paper, we propose a novel summarization approach with a flexible and reliable salience guidance, namely SEASON (SaliencE Allocation as Guidance for Abstractive SummarizatiON).SEASON utilizes the allocation of salience expectation to guide abstractive summarization and adapts well to articles in different abstractiveness.Automatic and human evaluations on two benchmark datasets show that the proposed method is effective and reliable.Empirical results on more than one million news articles demonstrate a natural fifteen-fifty salience split for news article sentences, providing a useful insight for composing news articles.

Fine-tuned Language Models are Continual Learners
Thomas Scialom | Tuhin Chakrabarty | Smaranda Muresan

Recent work on large language models relies on the intuition that most natural language processing tasks can be described via natural language instructions and that models trained on these instructions show strong zero-shot performance on several standard datasets. However, these models even though impressive still perform poorly on a wide range of tasks outside of their respective training and evaluation sets.To address this limitation, we argue that a model should be able to keep extending its knowledge and abilities, without forgetting previous skills. In spite of the limited success of Continual Learning, we show that Fine-tuned Language Models can be continual learners.We empirically investigate the reason for this success and conclude that Continual Learning emerges from self-supervision pre-training. Our resulting model Continual-T0 (CT0) is able to learn 8 new diverse language generation tasks, while still maintaining good performance on previous tasks, spanning in total of 70 datasets. Finally, we show that CT0 is able to combine instructions in ways it was never trained for, demonstrating some level of instruction compositionality.

Natural Logic-guided Autoregressive Multi-hop Document Retrieval for Fact Verification
Rami Aly | Andreas Vlachos

A key component of fact verification is the evidence retrieval, often from multiple documents. Recent approaches use dense representations and condition the retrieval of each document on the previously retrieved ones. The latter step is performed over all the documents in the collection, requiring storing their dense representations in an index, thus incurring a high memory footprint. An alternative paradigm is retrieve-and-rerank, where documents are retrieved using methods such as BM25, their sentences are reranked, and further documents are retrieved conditioned on these sentences, reducing the memory requirements. However, such approaches can be brittle as they rely on heuristics and assume hyperlinks between documents.We propose a novel retrieve-and-rerank method for multi-hop retrieval, that consists of a retriever that jointly scores documents in the knowledge source and sentences from previously retrieved documents using an autoregressive formulation and is guided by a proof system based on natural logic that dynamically terminates the retrieval process if the evidence is deemed sufficient.This method exceeds or is on par with the current state-of-the-art on FEVER, HoVer and FEVEROUS-S, while using 5 to 10 times less memory than competing systems. Evaluation on an adversarial dataset indicates improved stability of our approach compared to commonly deployed threshold-based methods. Finally, the proof system helps humans predict model decisions correctly more often than using the evidence alone.

AX-MABSA: A Framework for Extremely Weakly Supervised Multi-label Aspect Based Sentiment Analysis
Sabyasachi Kamila | Walid Magdy | Sourav Dutta | MingXue Wang

Aspect Based Sentiment Analysis is a dominant research area with potential applications in social media analytics, business, finance, and health. Prior works in this area are primarily based on supervised methods, with a few techniques using weak supervision limited to predicting a single aspect category per review sentence. In this paper, we present an extremely weakly supervised multi-label Aspect Category Sentiment Analysis framework which does not use any labelled data. We only rely on a single word per class as an initial indicative information. We further propose an automatic word selection technique to choose these seed categories and sentiment words. We explore unsupervised language model post-training to improve the overall performance, and propose a multi-label generator model to generate multiple aspect category-sentiment pairs per review sentence. Experiments conducted on four benchmark datasets showcase our method to outperform other weakly supervised baselines by a significant margin.

Transfer Learning with Synthetic Corpora for Spatial Role Labeling and Reasoning
Roshanak Mirzaee | Parisa Kordjamshidi

Recent research shows synthetic data as a source of supervision helps pretrained language models (PLM) transfer learning to new target tasks/domains. However, this idea is less explored for spatial language. We provide two new data resources on multiple spatial language processing tasks. The first dataset is synthesized for transfer learning on spatial question answering (SQA) and spatial role labeling (SpRL). Compared to previous SQA datasets, we include a larger variety of spatial relation types and spatial expressions. Our data generation process is easily extendable with new spatial expression lexicons. The second one is a real-world SQA dataset with human-generated questions built on an existing corpus with SPRL annotations. This dataset can be used to evaluate spatial language processing models in realistic situations. We show pretraining with automatically generated data significantly improves the SOTA results on several SQA and SPRL benchmarks, particularly when the training data in the target domain is small.

A Survey of Active Learning for Natural Language Processing
Zhisong Zhang | Emma Strubell | Eduard Hovy

In this work, we provide a literature review of active learning (AL) for its applications in natural language processing (NLP). In addition to a fine-grained categorization of query strategies, we also investigate several other important aspects of applying AL to NLP problems. These include AL for structured prediction tasks, annotation cost, model learning (especially with deep neural models), and starting and stopping AL. Finally, we conclude with a discussion of related topics and future directions.

Bernice: A Multilingual Pre-trained Encoder for Twitter
Alexandra DeLucia | Shijie Wu | Aaron Mueller | Carlos Aguirre | Philip Resnik | Mark Dredze

The language of Twitter differs significantly from that of other domains commonly included in large language model training. While tweets are typically multilingual and contain informal language, including emoji and hashtags, most pre-trained language models for Twitter are either monolingual, adapted from other domains rather than trained exclusively on Twitter, or are trained on a limited amount of in-domain Twitter data.We introduce Bernice, the first multilingual RoBERTa language model trained from scratch on 2.5 billion tweets with a custom tweet-focused tokenizer. We evaluate on a variety of monolingual and multilingual Twitter benchmarks, finding that our model consistently exceeds or matches the performance of a variety of models adapted to social media data as well as strong multilingual baselines, despite being trained on less data overall.We posit that it is more efficient compute- and data-wise to train completely on in-domain data with a specialized domain-specific tokenizer.

CEFR-Based Sentence Difficulty Annotation and Assessment
Yuki Arase | Satoru Uchida | Tomoyuki Kajiwara

Controllable text simplification is a crucial assistive technique for language learning and teaching. One of the primary factors hindering its advancement is the lack of a corpus annotated with sentence difficulty levels based on language ability descriptions. To address this problem, we created the CEFR-based Sentence Profile (CEFR-SP) corpus, containing 17k English sentences annotated with the levels based on the Common European Framework of Reference for Languages assigned by English-education professionals. In addition, we propose a sentence-level assessment model to handle unbalanced level distribution because the most basic and highly proficient sentences are naturally scarce. In the experiments in this study, our method achieved a macro-F1 score of 84.5% in the level assessment, thus outperforming strong baselines employed in readability assessment.

Simple Questions Generate Named Entity Recognition Datasets
Hyunjae Kim | Jaehyo Yoo | Seunghyun Yoon | Jinhyuk Lee | Jaewoo Kang

Recent named entity recognition (NER) models often rely on human-annotated datasets requiring the vast engagement of professional knowledge on the target domain and entities. This work introduces an ask-to-generate approach, which automatically generates NER datasets by asking simple natural language questions to an open-domain question answering system (e.g., “Which disease?”). Despite using fewer training resources, our models solely trained on the generated datasets largely outperform strong low-resource models by 19.5 F1 score across six popular NER benchmarks. Our models also show competitive performance with rich-resource models that additionally leverage in-domain dictionaries provided by domain experts. In few-shot NER, we outperform the previous best model by 5.2 F1 score on three benchmarks and achieve new state-of-the-art performance.

TemporalWiki: A Lifelong Benchmark for Training and Evaluating Ever-Evolving Language Models
Joel Jang | Seonghyeon Ye | Changho Lee | Sohee Yang | Joongbo Shin | Janghoon Han | Gyeonghun Kim | Minjoon Seo

Language Models (LMs) become outdated as the world changes; they often fail to perform tasks requiring recent factual information which was absent or different during training, a phenomenon called temporal misalignment. This is especially a challenging problem because the research community still lacks a coherent dataset for assessing the adaptability of LMs to frequently-updated knowledge corpus such as Wikipedia. To this end, we introduce TemporalWiki, a lifelong benchmark for ever-evolving LMs that utilizes the difference between consecutive snapshots of English Wikipedia and English Wikidata for training and evaluation, respectively. The benchmark hence allows researchers to periodically track an LM’s ability to retain previous knowledge and acquire updated/new knowledge at each point in time. We also find that training an LM on the diff data through continual learning methods achieves similar or better perplexity than on the entire snapshot in our benchmark with 12 times less computational cost, which verifies that factual knowledge in LMs can be safely updated with minimal training data via continual learning.

Bi-Directional Iterative Prompt-Tuning for Event Argument Extraction
Lu Dai | Bang Wang | Wei Xiang | Yijun Mo

Recently, prompt-tuning has attracted growing interests in event argument extraction (EAE). However, the existing prompt-tuning methods have not achieved satisfactory performance due to the lack of consideration of entity information. In this paper, we propose a bi-directional iterative prompt-tuning method for EAE, where the EAE task is treated as a cloze-style task to take full advantage of entity information and pre-trained language models (PLMs). Furthermore, our method explores event argument interactions by introducing the argument roles of contextual entities into prompt construction. Since template and verbalizer are two crucial components in a cloze-style prompt, we propose to utilize the role label semantic knowledge to construct a semantic verbalizer and design three kind of templates for the EAE task. Experiments on the ACE 2005 English dataset with standard and low-resource settings show that the proposed method significantly outperforms the peer state-of-the-art methods.

Learning Robust Representations for Continual Relation Extraction via Adversarial Class Augmentation
Peiyi Wang | Yifan Song | Tianyu Liu | Binghuai Lin | Yunbo Cao | Sujian Li | Zhifang Sui

Continual relation extraction (CRE) aims to continually learn new relations from a class-incremental data stream. CRE model usually suffers from catastrophic forgetting problem, i.e., the performance of old relations seriously degrades when the model learns new relations. Most previous work attributes catastrophic forgetting to the corruption of the learned representations as new relations come, with an implicit assumption that the CRE models have adequately learned the old relations. In this paper, through empirical studies we argue that this assumption may not hold, and an important reason for catastrophic forgetting is that the learned representations do not have good robustness against the appearance of analogous relations in the subsequent learning process. To address this issue, we encourage the model to learn more precise and robust representations through a simple yet effective adversarial class augmentation mechanism (ACA), which is easy to implement and model-agnostic.Experimental results show that ACA can consistently improve the performance of state-of-the-art CRE models on two popular benchmarks.

ConvFinQA: Exploring the Chain of Numerical Reasoning in Conversational Finance Question Answering
Zhiyu Chen | Shiyang Li | Charese Smiley | Zhiqiang Ma | Sameena Shah | William Yang Wang

With the recent advance in large pre-trained language models, researchers have achieved record performances in NLP tasks that mostly focus on language pattern matching. The community is experiencing the shift of the challenge from how to model language to the imitation of complex reasoning abilities like human beings. In this work, we investigate the application domain of finance that involves real-world, complex numerical reasoning. We propose a new large-scale dataset, ConvFinQA, aiming to study the chain of numerical reasoning in conversational question answering. Our dataset poses great challenge in modeling long-range, complex numerical reasoning paths in real-world conversations. We conduct comprehensive experiments and analyses with both the neural symbolic methods and the prompting-based methods, to provide insights into the reasoning mechanisms of these two divisions. We believe our new dataset should serve as a valuable resource to push forward the exploration of real-world, complex reasoning tasks as the next research focus. Our dataset and code is publicly available at

A Span-based Multimodal Variational Autoencoder for Semi-supervised Multimodal Named Entity Recognition
Baohang Zhou | Ying Zhang | Kehui Song | Wenya Guo | Guoqing Zhao | Hongbin Wang | Xiaojie Yuan

Multimodal named entity recognition (MNER) on social media is a challenging task which aims to extract named entities in free text and incorporate images to classify them into user-defined types. However, the annotation for named entities on social media demands a mount of human efforts. The existing semi-supervised named entity recognition methods focus on the text modal and are utilized to reduce labeling costs in traditional NER. However, the previous methods are not efficient for semi-supervised MNER. Because the MNER task is defined to combine the text information with image one and needs to consider the mismatch between the posted text and image. To fuse the text and image features for MNER effectively under semi-supervised setting, we propose a novel span-based multimodal variational autoencoder (SMVAE) model for semi-supervised MNER. The proposed method exploits modal-specific VAEs to model text and image latent features, and utilizes product-of-experts to acquire multimodal features. In our approach, the implicit relations between labels and multimodal features are modeled by multimodal VAE. Thus, the useful information of unlabeled data can be exploited in our method under semi-supervised setting. Experimental results on two benchmark datasets demonstrate that our approach not only outperforms baselines under supervised setting, but also improves MNER performance with less labeled data than existing semi-supervised methods.

R-TeaFor: Regularized Teacher-Forcing for Abstractive Summarization
Guan-Yu Lin | Pu-Jen Cheng

Teacher-forcing is widely used in training sequence generation models to improve sampling efficiency and to stabilize training. However, teacher-forcing is vulnerable to the exposure bias problem. Previous works have attempted to address exposure bias by modifying the training data to simulate model-generated results. Nevertheless, they do not consider the pairwise relationship between the original training data and the modified ones, which provides more information during training. Hence, we propose Regularized Teacher-Forcing (R-TeaFor) to utilize this relationship for better regularization. Empirically, our experiments show that R-TeaFor outperforms previous summarization state-of-the-art models, and the results can be generalized to different pre-trained models.

Modeling Consistency Preference via Lexical Chains for Document-level Neural Machine Translation
Xinglin Lyu | Junhui Li | Shimin Tao | Hao Yang | Ying Qin | Min Zhang

In this paper we aim to relieve the issue of lexical translation inconsistency for document-level neural machine translation (NMT) by modeling consistency preference for lexical chains, which consist of repeated words in a source-side document and provide a representation of the lexical consistency structure of the document. Specifically, we first propose lexical-consistency attention to capture consistency context among words in the same lexical chains. Then for each lexical chain we define and learn a consistency-tailored latent variable, which will guide the translation of corresponding sentences to enhance lexical translation consistency. Experimental results on Chinese→English and French→English document-level translation tasks show that our approach not only significantly improves translation performance in BLEU, but also substantially alleviates the problem of the lexical translation inconsistency.

Just Fine-tune Twice: Selective Differential Privacy for Large Language Models
Weiyan Shi | Ryan Shea | Si Chen | Chiyuan Zhang | Ruoxi Jia | Zhou Yu

Protecting large language models from privacy leakage is becoming increasingly crucial with their wide adoption in real-world products. Yet applying *differential privacy* (DP), a canonical notion with provable privacy guarantees for machine learning models, to those models remains challenging due to the trade-off between model utility and privacy loss. Utilizing the fact that sensitive information in language data tends to be sparse, Shi et al. (2021) formalized a DP notion extension called *Selective Differential Privacy* (SDP) to protect only the sensitive tokens defined by a policy function. However, their algorithm only works for RNN-based models. In this paper, we develop a novel framework, *Just Fine-tune Twice* (JFT), that achieves SDP for state-of-the-art large transformer-based models. Our method is easy to implement: it first fine-tunes the model with *redacted* in-domain data, and then fine-tunes it again with the *original* in-domain data using a private training mechanism. Furthermore, we study the scenario of imperfect implementation of policy functions that misses sensitive tokens and develop systematic methods to handle it. Experiments show that our method achieves strong utility compared to previous baselines. We also analyze the SDP privacy guarantee empirically with the canary insertion attack.

Factorizing Content and Budget Decisions in Abstractive Summarization of Long Documents
Marcio Fonseca | Yftah Ziser | Shay B. Cohen

We argue that disentangling content selection from the budget used to cover salient content improves the performance and applicability of abstractive summarizers. Our method, FactorSum, does this disentanglement by factorizing summarization into two steps through an energy function: (1) generation of abstractive summary views covering salient information in subsets of the input document (document views); (2) combination of these views into a final summary, following a budget and content guidance. This guidance may come from different sources, including from an advisor model such as BART or BigBird, or in oracle mode – from the reference. This factorization achieves significantly higher ROUGE scores on multiple benchmarks for long document summarization, namely PubMed, arXiv, and GovReport. Most notably, our model is effective for domain adaptation. When trained only on PubMed samples, it achieves a 46.29 ROUGE-1 score on arXiv, outperforming PEGASUS trained in domain by a large margin. Our experimental results indicate that the performance gains are due to more flexible budget adaptation and processing of shorter contexts provided by partial document views.

Open-Domain Sign Language Translation Learned from Online Video
Bowen Shi | Diane Brentari | Gregory Shakhnarovich | Karen Livescu

Existing work on sign language translation – that is, translation from sign language videos into sentences in a written language – has focused mainly on (1) data collected in a controlled environment or (2) data in a specific domain, which limits the applicability to real-world settings. In this paper, we introduce OpenASL, a large-scale American Sign Language (ASL) - English dataset collected from online video sites (e.g., YouTube).OpenASL contains 288 hours of ASL videos in multiple domains from over 200 signers and is the largest publicly available ASL translation dataset to date. To tackle the challenges of sign language translation in realistic settings and without glosses, we propose a set of techniques including sign search as a pretext task for pre-training and fusion of mouthing and handshape features. The proposed techniques produce consistent and large improvements in translation quality, over baseline models basedon prior work.

Improving Temporal Generalization of Pre-trained Language Models with Lexical Semantic Change
Zhaochen Su | Zecheng Tang | Xinyan Guan | Lijun Wu | Min Zhang | Juntao Li

Recent research has revealed that neural language models at scale suffer from poor temporal generalization capability, i.e., language model pre-trained on static data from past years performs worse over time on emerging data. Existing methods mainly perform continual training to mitigate such a misalignment. While effective to some extent but is far from being addressed on both the language modeling and downstream tasks. In this paper, we empirically observe that temporal generalization is closely affiliated with lexical semantic change, which is one of the essential phenomena of natural languages. Based on this observation, we propose a simple yet effective lexical-level masking strategy to post-train a converged language model. Experiments on two pre-trained language models, two different classification tasks, and four benchmark datasets demonstrate the effectiveness of our proposed method over existing temporal adaptation methods, i.e., continual training with new data. Our code is available at

ULN: Towards Underspecified Vision-and-Language Navigation
Weixi Feng | Tsu-Jui Fu | Yujie Lu | William Yang Wang

Vision-and-Language Navigation (VLN) is a task to guide an embodied agent moving to a target position using language instructions. Despite the significant performance improvement, the wide use of fine-grained instructions fails to characterize more practical linguistic variations in reality. To fill in this gap, we introduce a new setting, namely Underspecified vision-and-Language Navigation (ULN), and associated evaluation datasets. ULN evaluates agents using multi-level underspecified instructions instead of purely fine-grained or coarse-grained, which is a more realistic and general setting. As a primary step toward ULN, we propose a VLN framework that consists of a classification module, a navigation agent, and an Exploitation-to-Exploration (E2E) module. Specifically, we propose to learn Granularity Specific Sub-networks (GSS) for the agent to ground multi-level instructions with minimal additional parameters. Then, our E2E module estimates grounding uncertainty and conducts multi-step lookahead exploration to improve the success rate further. Experimental results show that existing VLN models are still brittle to multi-level language underspecification. Our framework is more robust and outperforms the baselines on ULN by ~10% relative success rate across all levels.

Federated Model Decomposition with Private Vocabulary for Text Classification
Zhuo Zhang | Xiangjing Hu | Lizhen Qu | Qifan Wang | Zenglin Xu

With the necessity of privacy protection, it becomes increasingly vital to train deep neural models in a federated learning manner for natural language processing (NLP) tasks. However, recent studies show eavesdroppers (i.e., dishonest servers) can still reconstruct the private input in federated learning (FL). Such a data reconstruction attack relies on the mappings between vocabulary and associated word embedding in NLP tasks, which are unfortunately less studied in current FL methods. In this paper, we propose a fedrated model decomposition method that protects the privacy of vocabularies, shorted as FEDEVOCAB. In FEDEVOCAB, each participant keeps the local embedding layer in the local device and detaches the local embedding parameters from federated aggregation. However, it is challenging to train an accurate NLP model when the private mappings are unknown and vary across participants in a cross-device FL setting. To address this problem, we further propose an adaptive updating technique to improve the performance of local models. Experimental results show that FEDEVOCAB maintains competitive performance and provides better privacy-preserving capacity compared to status quo methods.

ReCo: Reliable Causal Chain Reasoning via Structural Causal Recurrent Neural Networks
Kai Xiong | Xiao Ding | Zhongyang Li | Li Du | Ting Liu | Bing Qin | Yi Zheng | Baoxing Huai

Causal chain reasoning (CCR) is an essential ability for many decision-making AI systems, which requires the model to build reliable causal chains by connecting causal pairs. However, CCR suffers from two main transitive problems: threshold effect and scene drift. In other words, the causal pairs to be spliced may have a conflicting threshold boundary or scenario.To address these issues, we propose a novel Reliable Causal chain reasoning framework (ReCo), which introduces exogenous variables to represent the threshold and scene factors of each causal pair within the causal chain, and estimates the threshold and scene contradictions across exogenous variables via structural causal recurrent neural networks (SRNN). Experiments show that ReCo outperforms a series of strong baselines on both Chinese and English CCR datasets. Moreover, by injecting reliable causal chain knowledge distilled by ReCo, BERT can achieve better performances on four downstream causal-related tasks than BERT models enhanced by other kinds of knowledge.

Video Question Answering: Datasets, Algorithms and Challenges
Yaoyao Zhong | Wei Ji | Junbin Xiao | Yicong Li | Weihong Deng | Tat-Seng Chua

This survey aims to sort out the recent advances in video question answering (VideoQA) and point towards future directions. We firstly categorize the datasets into 1) normal VideoQA, multi-modal VideoQA and knowledge-based VideoQA, according to the modalities invoked in the question-answer pairs, or 2) factoid VideoQA and inference VideoQA, according to the technical challenges in comprehending the questions and deriving the correct answers. We then summarize the VideoQA techniques, including those mainly designed for Factoid QA (e.g., the early spatio-temporal attention-based methods and the recently Transformer-based ones) and those targeted at explicit relation and logic inference (e.g., neural modular networks, neural symbolic methods, and graph-structured methods). Aside from the backbone techniques, we delve into the specific models and find out some common and useful insights either for video modeling, question answering, or for cross-modal correspondence learning. Finally, we point out the research trend of studying beyond factoid VideoQA to inference VideoQA, as well as towards the robustness and interpretability. Additionally,