The concept of unsupervised universal sentence encoders has gained traction recently, wherein pre-trained models generate effective task-agnostic fixed-dimensional representations for phrases, sentences and paragraphs. Such methods are of varying complexity, from simple weighted-averages of word vectors to complex language-models based on bidirectional transformers. In this work we propose a novel technique to generate sentence-embeddings in an unsupervised fashion by projecting the sentences onto a fixed-dimensional manifold with the objective of preserving local neighbourhoods in the original space. To delineate such neighbourhoods we experiment with several set-distance metrics, including the recently proposed Word Mover’s distance, while the fixed-dimensional projection is achieved by employing a scalable and efficient manifold approximation method rooted in topological data analysis. We test our approach, which we term EMAP or Embeddings by Manifold Approximation and Projection, on six publicly available text-classification datasets of varying size and complexity. Empirical results show that our method consistently performs similar to or better than several alternative state-of-the-art approaches.
Multi-document question generation focuses on generating a question that covers the common aspect of multiple documents. Such a model is useful in generating clarifying options. However, a naive model trained only using the targeted (‘positive’) document set may generate too generic questions that cover a larger scope than delineated by the document set. To address this challenge, we introduce the contrastive learning strategy where given ‘positive’ and ‘negative’ sets of documents, we generate a question that is closely related to the ‘positive’ set but is far away from the ‘negative’ set. This setting allows generated questions to be more specific and related to the target document set. To generate such specific questions, we propose Multi-Source Coordinated Question Generator (MSCQG), a novel framework that includes a supervised learning (SL) stage and a reinforcement learning (RL) stage. In the SL stage, a single-document question generator is trained. In the RL stage, a coordinator model is trained to find optimal attention weights to align multiple single-document generators, by optimizing a reward designed to promote specificity of generated questions. We also develop an effective auxiliary objective, named Set-induced Contrastive Regularization (SCR) that improves the coordinator’s contrastive learning during the RL stage. We show that our model significantly outperforms several strong baselines, as measured by automatic metrics and human evaluation. The source repository is publicly available at ‘www.github.com/woonsangcho/contrast_qgen’.
Psycholinguistic studies of human word processing and lexical access provide ample evidence of the preferred nature of word-initial versus word-final segments, e.g., in terms of attention paid by listeners (greater) or the likelihood of reduction by speakers (lower). This has led to the conjecture—as in Wedel et al. (2019b), but common elsewhere—that languages have evolved to provide more information earlier in words than later. Information-theoretic methods to establish such tendencies in lexicons have suffered from several methodological shortcomings that leave open the question of whether this high word-initial informativeness is actually a property of the lexicon or simply an artefact of the incremental nature of recognition. In this paper, we point out the confounds in existing methods for comparing the informativeness of segments early in the word versus later in the word, and present several new measures that avoid these confounds. When controlling for these confounds, we still find evidence across hundreds of languages that indeed there is a cross-linguistic tendency to front-load information in words.
Images are core components of multi-modal learning in natural language processing (NLP), and results have varied substantially as to whether images improve NLP tasks or not. One confounding effect has been that previous NLP research has generally focused on sophisticated tasks (in varying settings), generally applied to English only. We focus on text classification, in the context of assigning named entity classes to a given Wikipedia page, where images generally complement the text and the Wikipedia page can be in one of a number of different languages. Our experiments across a range of languages show that images complement NLP models (including BERT) trained without external pre-training, but when combined with BERT models pre-trained on large-scale external data, images contribute nothing.
Fine-grained sentiment analysis attempts to extract sentiment holders, targets and polar expressions and resolve the relationship between them, but progress has been hampered by the difficulty of annotation. Targeted sentiment analysis, on the other hand, is a more narrow task, focusing on extracting sentiment targets and classifying their polarity. In this paper, we explore whether incorporating holder and expression information can improve target extraction and classification and perform experiments on eight English datasets. We conclude that jointly predicting target and polarity BIO labels improves target extraction, and that augmenting the input text with gold expressions generally improves targeted polarity classification. This highlights the potential importance of annotating expressions for fine-grained sentiment datasets. At the same time, our results show that performance of current models for predicting polar expressions is poor, hampering the benefit of this information in practice.
A common approach in many machine learning algorithms involves self-supervised learning on large unlabeled data before fine-tuning on downstream tasks to further improve performance. A new approach for language modelling, called dynamic evaluation, further fine-tunes a trained model during inference using trivially-present ground-truth labels, giving a large improvement in performance. However, this approach does not easily extend to classification tasks, where ground-truth labels are absent during inference. We propose to solve this issue by utilizing self-training and back-propagating the loss from the model’s own class-balanced predictions (pseudo-labels), adapting the Reptile algorithm from meta-learning, combined with an inductive bias towards pre-trained weights to improve generalization. Our method improves the performance of standard backbones such as BERT, Electra, and ResNet-50 on a wide variety of tasks, such as question answering on SQuAD and NewsQA, benchmark task SuperGLUE, conversation response selection on Ubuntu Dialog corpus v2.0, as well as image classification on MNIST and ImageNet without any changes to the underlying models. Our proposed method outperforms previous approaches, enables self-supervised fine-tuning during inference of any classifier model to better adapt to target domains, can be easily adapted to any model, and is also effective in online and transfer-learning settings.
Modelling persuasion strategies as predictors of task outcome has several real-world applications and has received considerable attention from the computational linguistics community. However, previous research has failed to account for the resisting strategies employed by an individual to foil such persuasion attempts. Grounded in prior literature in cognitive and social psychology, we propose a generalised framework for identifying resisting strategies in persuasive conversations. We instantiate our framework on two distinct datasets comprising persuasion and negotiation conversations. We also leverage a hierarchical sequence-labelling neural architecture to infer the aforementioned resisting strategies automatically. Our experiments reveal the asymmetry of power roles in non-collaborative goal-directed conversations and the benefits accrued from incorporating resisting strategies on the final conversation outcome. We also investigate the role of different resisting strategies on the conversation outcome and glean insights that corroborate with past findings. We also make the code and the dataset of this work publicly available at https://github.com/americast/resper.
The slow speed of BERT has motivated much research on accelerating its inference, and the early exiting idea has been proposed to make trade-offs between model quality and efficiency. This paper aims to address two weaknesses of previous work: (1) existing fine-tuning strategies for early exiting models fail to take full advantage of BERT; (2) methods to make exiting decisions are limited to classification tasks. We propose a more advanced fine-tuning strategy and a learning-to-exit module that extends early exiting to tasks other than classification. Experiments demonstrate improved early exiting for BERT, with better trade-offs obtained by the proposed fine-tuning strategy, successful application to regression tasks, and the possibility to combine it with other acceleration methods. Source code can be found at https://github.com/castorini/berxit.
We take a deep look into the behaviour of self-attention heads in the transformer architecture. In light of recent work discouraging the use of attention distributions for explaining a model’s behaviour, we show that attention distributions can nevertheless provide insights into the local behaviour of attention heads. This way, we propose a distinction between local patterns revealed by attention and global patterns that refer back to the input, and analyze BERT from both angles. We use gradient attribution to analyze how the output of an attention head depends on the input tokens, effectively extending the local attention-based analysis to account for the mixing of information throughout the transformer layers. We find that there is a significant mismatch between attention and attribution distributions, caused by the mixing of context inside the model. We quantify this discrepancy and observe that interestingly, there are some patterns that persist across all layers despite the mixing.
Lexical semantic change detection is a new and innovative research field. The optimal fine-tuning of models including pre- and post-processing is largely unclear. We optimize existing models by (i) pre-training on large corpora and refining on diachronic target corpora tackling the notorious small data problem, and (ii) applying post-processing transformations that have been shown to improve performance on synchronic tasks. Our results provide a guide for the application and optimization of lexical semantic change detection models across various learning scenarios.
Large datasets are essential for neural modeling of many NLP tasks. Current publicly available open-domain dialogue datasets offer a trade-off between quality (e.g., DailyDialog) and size (e.g., Opensubtitles). We narrow this gap by building a high-quality dataset of 14.8M utterances in English, and smaller datasets in German, Dutch, Spanish, Portuguese, Italian, and Hungarian. We extract and process dialogues from public-domain books made available by Project Gutenberg. We describe our dialogue extraction pipeline, analyze the effects of the various heuristics used, and present an error analysis of extracted dialogues. Finally, we conduct experiments showing that better response quality can be achieved in zero-shot and finetuning settings by training on our data than on the larger but much noisier Opensubtitles dataset. Our open-source pipeline (https://github.com/ricsinaruto/gutenberg-dialog) can be extended to further languages with little additional effort. Researchers can also build their versions of existing datasets by adjusting various trade-off parameters.
According to the Probability Ranking Principle (PRP), ranking documents in decreasing order of their probability of relevance leads to an optimal document ranking for ad-hoc retrieval. The PRP holds when two conditions are met: [C1] the models are well calibrated, and, [C2] the probabilities of relevance are reported with certainty. We know however that deep neural networks (DNNs) are often not well calibrated and have several sources of uncertainty, and thus [C1] and [C2] might not be satisfied by neural rankers. Given the success of neural Learning to Rank (LTR) approaches—and here, especially BERT-based approaches—we first analyze under which circumstances deterministic neural rankers are calibrated for conversational search problems. Then, motivated by our findings we use two techniques to model the uncertainty of neural rankers leading to the proposed stochastic rankers, which output a predictive distribution of relevance as opposed to point estimates. Our experimental results on the ad-hoc retrieval task of conversation response ranking reveal that (i) BERT-based rankers are not robustly calibrated and that stochastic BERT-based rankers yield better calibration; and (ii) uncertainty estimation is beneficial for both risk-aware neural ranking, i.e. taking into account the uncertainty when ranking documents, and for predicting unanswerable conversational contexts.
Recent efforts have shown that neural text processing models are vulnerable to adversarial examples, but the nature of these examples is poorly understood. In this work, we show that adversarial attacks against CNN, LSTM and Transformer-based classification models perform word substitutions that are identifiable through frequency differences between replaced words and their corresponding substitutions. Based on these findings, we propose frequency-guided word substitutions (FGWS), a simple algorithm exploiting the frequency properties of adversarial word substitutions for the detection of adversarial examples. FGWS achieves strong performance by accurately detecting adversarial examples on the SST-2 and IMDb sentiment datasets, with F1 detection scores of up to 91.4% against RoBERTa-based classification models. We compare our approach against a recently proposed perturbation discrimination framework and show that we outperform it by up to 13.0% F1.
Language modeling with BERT consists of two phases of (i) unsupervised pre-training on unlabeled text, and (ii) fine-tuning for a specific supervised task. We present a method that leverages the second phase to its fullest, by applying an extensive number of parallel classifier heads, which are enforced to be orthogonal, while adaptively eliminating the weaker heads during training. We conduct an extensive inter- and intra-dataset evaluation, showing that our method improves the generalization ability of BERT, sometimes leading to a +9% gain in accuracy. These results highlight the importance of a proper fine-tuning procedure, especially for relatively smaller-sized datasets. Our code is attached as supplementary.
This paper presents a novel framework for reconstructing multi-hop explanations in science Question Answering (QA). While existing approaches for multi-hop reasoning build explanations considering each question in isolation, we propose a method to leverage explanatory patterns emerging in a corpus of scientific explanations. Specifically, the framework ranks a set of atomic facts by integrating lexical relevance with the notion of unification power, estimated analysing explanations for similar questions in the corpus. An extensive evaluation is performed on the Worldtree corpus, integrating k-NN clustering and Information Retrieval (IR) techniques. We present the following conclusions: (1) The proposed method achieves results competitive with Transformers, yet being orders of magnitude faster, a feature that makes it scalable to large explanatory corpora (2) The unification-based mechanism has a key role in reducing semantic drift, contributing to the reconstruction of many hops explanations (6 or more facts) and the ranking of complex inference facts (+12.0 Mean Average Precision) (3) Crucially, the constructed explanations can support downstream QA models, improving the accuracy of BERT by up to 10% overall.
Word embeddings trained on large corpora have shown to encode high levels of unfair discriminatory gender, racial, religious and ethnic biases. In contrast, human-written dictionaries describe the meanings of words in a concise, objective and an unbiased manner. We propose a method for debiasing pre-trained word embeddings using dictionaries, without requiring access to the original training resources or any knowledge regarding the word embedding algorithms used. Unlike prior work, our proposed method does not require the types of biases to be pre-defined in the form of word lists, and learns the constraints that must be satisfied by unbiased word embeddings automatically from dictionary definitions of the words. Specifically, we learn an encoder to generate a debiased version of an input word embedding such that it (a) retains the semantics of the pre-trained word embedding, (b) agrees with the unbiased definition of the word according to the dictionary, and (c) remains orthogonal to the vector space spanned by any biased basis vectors in the pre-trained word embedding space. Experimental results on standard benchmark datasets show that the proposed method can accurately remove unfair biases encoded in pre-trained word embeddings, while preserving useful semantics.
In this work, we argue that augmenting argument generation technology with the ability to encode beliefs is of twofold. First, it gives more control on the generated arguments leading to better reach for audience. Second, it is one way of modeling the human process of synthesizing arguments. Therefore, we propose the task of belief-based claim generation, and study the research question of how to model and encode a user’s beliefs into a generated argumentative text. To this end, we model users’ beliefs via their stances on big issues, and extend state of the art text generation models with extra input reflecting user’s beliefs. Through an automatic evaluation we show empirical evidence of the applicability to encode beliefs into argumentative text. In our manual evaluation, we highlight that the low effectiveness of our approach stems from the noise produced by the automatic collection of bag-of-words, which was mitigated by removing this noise. The finding of this paper lays the ground work to further investigate the role of beliefs in generating better reaching arguments.
Non-autoregressive generation (NAG) has recently attracted great attention due to its fast inference speed. However, the generation quality of existing NAG models still lags behind their autoregressive counterparts. In this work, we show that BERT can be employed as the backbone of a NAG model for a greatly improved performance. Additionally, we devise two mechanisms to alleviate the two common problems of vanilla NAG models: the inflexibility of prefixed output length and the conditional independence of individual token predictions. To further strengthen the speed advantage of the proposed model, we propose a new decoding strategy, ratio-first, for applications where the output lengths can be approximately estimated beforehand. For a comprehensive evaluation, we test the proposed model on three text generation tasks, including text summarization, sentence compression and machine translation. Experimental results show that our model significantly outperforms existing non-autoregressive baselines and achieves competitive performance with many strong autoregressive models. In addition, we also conduct extensive analysis experiments to reveal the effect of each proposed component.
Large-scale transformers have been shown the state-of-the-art on neural machine translation. However, training these increasingly wider and deeper models could be tremendously memory intensive. We reduce the memory burden by employing the idea of reversible networks that a layer’s input can be reconstructed from its output. We design three types of multi-split based reversible transformers. We also devise a corresponding backpropagation algorithm, which does not need to store activations for most layers. Furthermore, we present two fine-tuning techniques: splits shuffle and self ensemble, to boost translation accuracy. Specifically, our best models surpass the vanilla transformer by at least 1.4 BLEU points in three datasets. Our large-scale reversible models achieve 30.0 BLEU in WMT’14 En-De and 43.5 BLEU in WMT’14 En-Fr, beating several very strong baselines with less than half of the training memory.
Some NLP tasks can be solved in a fully unsupervised fashion by providing a pretrained language model with “task descriptions” in natural language (e.g., Radford et al., 2019). While this approach underperforms its supervised counterpart, we show in this work that the two ideas can be combined: We introduce Pattern-Exploiting Training (PET), a semi-supervised training procedure that reformulates input examples as cloze-style phrases to help language models understand a given task. These phrases are then used to assign soft labels to a large set of unlabeled examples. Finally, standard supervised training is performed on the resulting training set. For several tasks and languages, PET outperforms supervised training and strong semi-supervised approaches in low-resource settings by a large margin.
Cross-document co-reference resolution (CDCR) is the task of identifying and linking mentions to entities and concepts across many text documents. Current state-of-the-art models for this task assume that all documents are of the same type (e.g. news articles) or fall under the same theme. However, it is also desirable to perform CDCR across different domains (type or theme). A particular use case we focus on in this paper is the resolution of entities mentioned across scientific work and newspaper articles that discuss them. Identifying the same entities and corresponding concepts in both scientific articles and news can help scientists understand how their work is represented in mainstream media. We propose a new task and English language dataset for cross-document cross-domain co-reference resolution (CDˆ2CR). The task aims to identify links between entities across heterogeneous document types. We show that in this cross-domain, cross-document setting, existing CDCR models do not perform well and we provide a baseline model that outperforms current state-of-the-art CDCR models on CDˆ2CR. Our data set, annotation tool and guidelines as well as our model for cross-document cross-domain co-reference are all supplied as open access open source resources.
Redundancy-aware extractive summarization systems score the redundancy of the sentences to be included in a summary either jointly with their salience information or separately as an additional sentence scoring step. Previous work shows the efficacy of jointly scoring and selecting sentences with neural sequence generation models. It is, however, not well-understood if the gain is due to better encoding techniques or better redundancy reduction approaches. Similarly, the contribution of salience versus diversity components on the created summary is not studied well. Building on the state-of-the-art encoding methods for summarization, we present two adaptive learning models: AREDSUM-SEQ that jointly considers salience and novelty during sentence selection; and a two-step AREDSUM-CTX that scores salience first, then learns to balance salience and redundancy, enabling the measurement of the impact of each aspect. Empirical results on CNN/DailyMail and NYT50 datasets show that by modeling diversity explicitly in a separate step, AREDSUM-CTX achieves significantly better performance than AREDSUM-SEQ as well as state-of-the-art extractive summarization baselines.
It has been well-documented for several languages that human interlocutors tend to adapt their linguistic productions to become more similar to each other. This behavior, known as entrainment, affects lexical choice as well, both with regard to specific words, such as referring expressions, and overall style. We offer what we believe to be the first investigation of such lexical entrainment in Hebrew. Using two existing measures, we analyze Hebrew speakers interacting in a Map Task, a popular experimental setup, and find rich evidence of lexical entrainment. Analyzing speaker pairs by the combination of their genders as well as speakers by their individual gender, we find no clear pattern of differences. We do, however, find that speakers in a position of less power entrain more than those with greater power, which matches theoretical accounts. Overall, our results mostly accord with those for American English, with a lack of entrainment on hedge words being the main difference.
Building open-domain chatbots is a challenging area for machine learning research. While prior work has shown that scaling neural models in the number of parameters and the size of the data they are trained on gives improved results, we highlight other ingredients. Good conversation requires blended skills: providing engaging talking points, and displaying knowledge, empathy and personality appropriately, while maintaining a consistent persona. We show that large scale models can learn these skills when given appropriate training data and choice of generation strategy. We build variants of these recipes with 90M, 2.7B and 9.4B parameter models, and make our models and code publicly available. Human evaluations show our best models outperform existing approaches in multi-turn dialogue on engagingness and humanness measurements. We then discuss the limitations of this work by analyzing failure cases of our models.
Despite growing interest in natural language generation (NLG) models that produce diverse outputs, there is currently no principled method for evaluating the diversity of an NLG system. In this work, we propose a framework for evaluating diversity metrics. The framework measures the correlation between a proposed diversity metric and a diversity parameter, a single parameter that controls some aspect of diversity in generated text. For example, a diversity parameter might be a binary variable used to instruct crowdsourcing workers to generate text with either low or high content diversity. We demonstrate the utility of our framework by: (a) establishing best practices for eliciting diversity judgments from humans, (b) showing that humans substantially outperform automatic metrics in estimating content diversity, and (c) demonstrating that existing methods for controlling diversity by tuning a “decoding parameter” mostly affect form but not meaning. Our framework can advance the understanding of different diversity metrics, an essential step on the road towards better NLG systems.
Question answering over knowledge bases (KBQA) usually involves three sub-tasks, namely topic entity detection, entity linking and relation detection. Due to the large number of entities and relations inside knowledge bases (KB), previous work usually utilized sophisticated rules to narrow down the search space and managed only a subset of KBs in memory. In this work, we leverage a retrieve-and-rerank framework to access KBs via traditional information retrieval (IR) method, and re-rank retrieved candidates with more powerful neural networks such as the pre-trained BERT model. Considering the fact that directly assigning a different BERT model for each sub-task may incur prohibitive costs, we propose to share a BERT encoder across all three sub-tasks and define task-specific layers on top of the shared layer. The unified model is then trained under a multi-task learning framework. Experiments show that: (1) Our IR-based retrieval method is able to collect high-quality candidates efficiently, thus enables our method adapt to large-scale KBs easily; (2) the BERT model improves the accuracy across all three sub-tasks; and (3) benefiting from multi-task learning, the unified model obtains further improvements with only 1/3 of the original parameters. Our final model achieves competitive results on the SimpleQuestions dataset and superior performance on the FreebaseQA dataset.
We examine the task of detecting implicitly abusive comparisons (e.g. “Your hair looks like you have been electrocuted”). Implicitly abusive comparisons are abusive comparisons in which abusive words (e.g. “dumbass” or “scum”) are absent. We detail the process of creating a novel dataset for this task via crowdsourcing that includes several measures to obtain a sufficiently representative and unbiased set of comparisons. We also present classification experiments that include a range of linguistic features that help us better understand the mechanisms underlying abusive comparisons.
We propose to use abusive emojis, such as the “middle finger” or “face vomiting”, as a proxy for learning a lexicon of abusive words. Since it represents extralinguistic information, a single emoji can co-occur with different forms of explicitly abusive utterances. We show that our approach generates a lexicon that offers the same performance in cross-domain classification of abusive microposts as the most advanced lexicon induction method. Such an approach, in contrast, is dependent on manually annotated seed words and expensive lexical resources for bootstrapping (e.g. WordNet). We demonstrate that the same emojis can also be effectively used in languages other than English. Finally, we also show that emojis can be exploited for classifying mentions of ambiguous words, such as “fuck” and “bitch”, into generally abusive and just profane usages.
Against the background of what has been termed a reproducibility crisis in science, the NLP field is becoming increasingly interested in, and conscientious about, the reproducibility of its results. The past few years have seen an impressive range of new initiatives, events and active research in the area. However, the field is far from reaching a consensus about how reproducibility should be defined, measured and addressed, with diversity of views currently increasing rather than converging. With this focused contribution, we aim to provide a wide-angle, and as near as possible complete, snapshot of current work on reproducibility in NLP,
We develop high performance multilingual Abstract Meaning Representation (AMR) systems by projecting English AMR annotations to other languages with weak supervision. We achieve this goal by bootstrapping transformer-based multilingual word embeddings, in particular those from cross-lingual RoBERTa (XLM-R large). We develop a novel technique for foreign-text-to-English AMR alignment, using the contextual word alignment between English and foreign language tokens. This word alignment is weakly supervised and relies on the contextualized XLM-R word embeddings. We achieve a highly competitive performance that surpasses the best published results for German, Italian, Spanish and Chinese.
Social networks face a major challenge in the form of rumors and fake news, due to their intrinsic nature of connecting users to millions of others, and of giving any individual the power to post anything. Given the rapid, widespread dissemination of information in social networks, manually detecting suspicious news is sub-optimal. Thus, research on automatic rumor detection has become a necessity. Previous works in the domain have utilized the reply relations between posts, as well as the semantic similarity between the main post and its context, consisting of replies, in order to obtain state-of-the-art performance. In this work, we demonstrate that semantic oppositeness can improve the performance on the task of rumor detection. We show that semantic oppositeness captures elements of discord, which are not properly covered by previous efforts, which only utilize semantic similarity or reply structure. We show, with extensive experiments on recent data sets for this problem, that our proposed model achieves state-of-the-art performance. Further, we show that our model is more resistant to the variances in performance introduced by randomness.
Learning disentangled representations of realworld data is a challenging open problem. Most previous methods have focused on either supervised approaches which use attribute labels or unsupervised approaches that manipulate the factorization in the latent space of models such as the variational autoencoder (VAE) by training with task-specific losses. In this work, we propose polarized-VAE, an approach that disentangles select attributes in the latent space based on proximity measures reflecting the similarity between data points with respect to these attributes. We apply our method to disentangle the semantics and syntax of sentences and carry out transfer experiments. Polarized-VAE outperforms the VAE baseline and is competitive with state-of-the-art approaches, while being more a general framework that is applicable to other attribute disentanglement tasks.
We propose ParaSCI, the first large-scale paraphrase dataset in the scientific field, including 33,981 paraphrase pairs from ACL (ParaSCI-ACL) and 316,063 pairs from arXiv (ParaSCI-arXiv). Digging into characteristics and common patterns of scientific papers, we construct this dataset though intra-paper and inter-paper methods, such as collecting citations to the same paper or aggregating definitions by scientific terms. To take advantage of sentences paraphrased partially, we put up PDBERT as a general paraphrase discovering method. The major advantages of paraphrases in ParaSCI lie in the prominent length and textual diversity, which is complementary to existing paraphrase datasets. ParaSCI obtains satisfactory results on human evaluation and downstream tasks, especially long paraphrase generation.
We introduce a general framework for abstractive summarization with factual consistency and distinct modeling of the narrative flow in an output summary. Our work addresses current limitations of models for abstractive summarization that often hallucinate information or generate summaries with coherence issues. To generate abstractive summaries with factual consistency and narrative flow, we propose Cooperative Generator-Discriminator Networks (Co-opNet), a novel transformer-based framework where the generator works with a discriminator architecture to compose coherent long-form summaries. We explore four different discriminator objectives which each capture a different aspect of coherence, including whether salient spans of generated abstracts are hallucinated or appear in the input context, and the likelihood of sentence adjacency in generated abstracts. We measure the ability of Co-opNet to learn these objectives with arXiv scientific papers, using the abstracts as a proxy for gold long-form scientific article summaries. Empirical results from automatic and human evaluations demonstrate that Co-opNet learns to summarize with considerably improved global coherence compared to competitive baselines.
Knowledge Base Question Answering (KBQA) is the problem of predicting an answer for a factoid question over a given knowledge base (KB). Answering questions typically requires reasoning over multiple links in the given KB. Humans tend to answer questions by grouping different objects to perform reasoning over acquired knowledge. Hypergraphs provide a natural tool to model group relationships. In this work, inspired by typical human intelligence, we propose a new method for KBQA based on hypergraphs. Existing methods for KBQA, though effective, do not explicitly incorporate the recursive relational group structure in the given KB. Our method, which we name RecHyperNet (Recursive Hypergraph Network), exploits a new way of modelling KBs through recursive hypergraphs to organise such group relationships in KBs. Experiments on multiple KBQA benchmarks demonstrate the effectiveness of the proposed RecHyperNet. We have released the code.
Current models for Word Sense Disambiguation (WSD) struggle to disambiguate rare senses, despite reaching human performance on global WSD metrics. This stems from a lack of data for both modeling and evaluating rare senses in existing WSD datasets. In this paper, we introduce FEWS (Few-shot Examples of Word Senses), a new low-shot WSD dataset automatically extracted from example sentences in Wiktionary. FEWS has high sense coverage across different natural language domains and provides: (1) a large training set that covers many more senses than previous datasets and (2) a comprehensive evaluation set containing few- and zero-shot examples of a wide variety of senses. We establish baselines on FEWS with knowledge-based and neural WSD approaches and present transfer learning experiments demonstrating that models additionally trained with FEWS better capture rare senses in existing WSD datasets. Finally, we find humans outperform the best baseline models on FEWS, indicating that FEWS will support significant future work on low-shot WSD.
In conversational analyses, humans manually weave multimodal information into the transcripts, which is significantly time-consuming. We introduce a system that automatically expands the verbatim transcripts of video-recorded conversations using multimodal data streams. This system uses a set of preprocessing rules to weave multimodal annotations into the verbatim transcripts and promote interpretability. Our feature engineering contributions are two-fold: firstly, we identify the range of multimodal features relevant to detect rapport-building; secondly, we expand the range of multimodal annotations and show that the expansion leads to statistically significant improvements in detecting rapport-building.
Bridging the performance gap between high- and low-resource languages has been the focus of much previous work. Typological features from databases such as the World Atlas of Language Structures (WALS) are a prime candidate for this, as such data exists even for very low-resource languages. However, previous work has only found minor benefits from using typological information. Our hypothesis is that a model trained in a cross-lingual setting will pick up on typological cues from the input data, thus overshadowing the utility of explicitly using such features. We verify this hypothesis by blinding a model to typological information, and investigate how cross-lingual sharing and performance is impacted. Our model is based on a cross-lingual architecture in which the latent weights governing the sharing between languages is learnt during training. We show that (i) preventing this model from exploiting typology severely reduces performance, while a control experiment reaffirms that (ii) encouraging sharing according to typology somewhat improves performance.
Sequential fine-tuning and multi-task learning are methods aiming to incorporate knowledge from multiple tasks; however, they suffer from catastrophic forgetting and difficulties in dataset balancing. To address these shortcomings, we propose AdapterFusion, a new two stage learning algorithm that leverages knowledge from multiple tasks. First, in the knowledge extraction stage we learn task specific parameters called adapters, that encapsulate the task-specific information. We then combine the adapters in a separate knowledge composition step. We show that by separating the two stages, i.e., knowledge extraction and knowledge composition, the classifier can effectively exploit the representations learned from multiple tasks in a non-destructive manner. We empirically evaluate AdapterFusion on 16 diverse NLU tasks, and find that it effectively combines various types of knowledge at different layers of the model. We show that our approach outperforms traditional strategies such as full fine-tuning as well as multi-task learning. Our code and adapters are available at AdapterHub.ml.
In this paper, we propose CHOLAN, a modular approach to target end-to-end entity linking (EL) over knowledge bases. CHOLAN consists of a pipeline of two transformer-based models integrated sequentially to accomplish the EL task. The first transformer model identifies surface forms (entity mentions) in a given text. For each mention, a second transformer model is employed to classify the target entity among a predefined candidates list. The latter transformer is fed by an enriched context captured from the sentence (i.e. local context), and entity description gained from Wikipedia. Such external contexts have not been used in state of the art EL approaches. Our empirical study was conducted on two well-known knowledge bases (i.e., Wikidata and Wikipedia). The empirical results suggest that CHOLAN outperforms state-of-the-art approaches on standard datasets such as CoNLL-AIDA, MSNBC, AQUAINT, ACE2004, and T-REx.
Collaborative grounding is a fundamental aspect of human-human dialog which allows people to negotiate meaning. In this paper we argue that it is missing from current deep learning approaches to dialog. Our central point is that making mistakes and being able to recover from them collaboratively is a key ingredient in grounding meaning. We illustrate the pitfalls of being unable to ground collaboratively, discuss what can be learned from the language acquisition and dialog systems literature, and reflect on how to move forward.
Recent progress in pretraining language models on large corpora has resulted in significant performance gains on many NLP tasks. These large models acquire linguistic knowledge during pretraining, which helps to improve performance on downstream tasks via fine-tuning. To assess what kind of knowledge is acquired, language models are commonly probed by querying them with ‘fill in the blank’ style cloze questions. Existing probing datasets mainly focus on knowledge about relations between words and entities. We introduce WDLMPro (Word Definitions Language Model Probing) to evaluate word understanding directly using dictionary definitions of words. In our experiments, three popular pretrained language models struggle to match words and their definitions. This indicates that they understand many words poorly and that our new probing task is a difficult challenge that could help guide research on LMs in the future.
Character linking, the task of linking mentioned people in conversations to the real world, is crucial for understanding the conversations. For the efficiency of communication, humans often choose to use pronouns (e.g., “she”) or normal entities (e.g., “that girl”) rather than named entities (e.g., “Rachel”) in the spoken language, which makes linking those mentions to real people a much more challenging than a regular entity linking task. To address this challenge, we propose to incorporate the richer context from the coreference relations among different mentions to help the linking. On the other hand, considering that finding coreference clusters itself is not a trivial task and could benefit from the global character information, we propose to jointly solve these two tasks. Specifically, we propose Cˆ2, the joint learning model of Coreference resolution and Character linking. The experimental results demonstrate that Cˆ2 can significantly outperform previous works on both tasks. Further analyses are conducted to analyze the contribution of all modules in the proposed model and the effect of all hyper-parameters.
Neural models for response generation produce responses that are semantically plausible but not necessarily factually consistent with facts describing the speaker’s persona. These models are trained with fully supervised learning where the objective function barely captures factual consistency. We propose to fine-tune these models by reinforcement learning and an efficient reward function that explicitly captures the consistency between a response and persona facts as well as semantic plausibility. Our automatic and human evaluations on the PersonaChat corpus confirm that our approach increases the rate of responses that are factually consistent with persona facts over its supervised counterpart while retains the language quality of responses.
To avoid the “meaning conflation deficiency” of word embeddings, a number of models have aimed to embed individual word senses. These methods at one time performed well on tasks such as word sense induction (WSI), but they have since been overtaken by task-specific techniques which exploit contextualized embeddings. However, sense embeddings and contextualization need not be mutually exclusive. We introduce PolyLM, a method which formulates the task of learning sense embeddings as a language modeling problem, allowing contextualization techniques to be applied. PolyLM is based on two underlying assumptions about word senses: firstly, that the probability of a word occurring in a given context is equal to the sum of the probabilities of its individual senses occurring; and secondly, that for a given occurrence of a word, one of its senses tends to be much more plausible in the context than the others. We evaluate PolyLM on WSI, showing that it performs considerably better than previous sense embedding techniques, and matches the current state-of-the-art specialized WSI method despite having six times fewer parameters. Code and pre-trained models are available at https://github.com/AlanAnsell/PolyLM.
We investigate the feasibility of applying standard text categorisation methods to patient text in order to predict treatment outcome in Internet-based cognitive behavioural therapy. The data set is unique in its detail and size for regular care for depression, social anxiety, and panic disorder. Our results indicate that there is a signal in the depression data, albeit a weak one. We also perform terminological and sentiment analysis, which confirm those results.
We consider the situation in which a user has collected a small set of documents on a cohesive topic, and they want to retrieve additional documents on this topic from a large collection. Information Retrieval (IR) solutions treat the document set as a query, and look for similar documents in the collection. We propose to extend the IR approach by treating the problem as an instance of positive-unlabeled (PU) learning—i.e., learning binary classifiers from only positive (the query documents) and unlabeled (the results of the IR engine) data. Utilizing PU learning for text with big neural networks is a largely unexplored field. We discuss various challenges in applying PU learning to the setting, showing that the standard implementations of state-of-the-art PU solutions fail. We propose solutions for each of the challenges and empirically validate them with ablation tests. We demonstrate the effectiveness of the new method using a series of experiments of retrieving PubMed abstracts adhering to fine-grained topics, showing improvements over the common IR solution and other baselines.
Image captioning has focused on generalizing to images drawn from the same distribution as the training set, and not to the more challenging problem of generalizing to different distributions of images. Recently, Nikolaus et al. (2019) introduced a dataset to assess compositional generalization in image captioning, where models are evaluated on their ability to describe images with unseen adjective–noun and noun–verb compositions. In this work, we investigate different methods to improve compositional generalization by planning the syntactic structure of a caption. Our experiments show that jointly modeling tokens and syntactic tags enhances generalization in both RNN- and Transformer-based models, while also improving performance on standard metrics.
Modelling a word’s polarity in different contexts is a key task in sentiment analysis. Previous works mainly focus on domain dependencies, and assume words’ sentiments are invariant within a specific domain. In this paper, we relax this assumption by binding a word’s sentiment to its collocation words instead of domain labels. This finer view of sentiment contexts is particularly useful for identifying commonsense sentiments expressed in neural words such as “big” and “long”. Given a target (e.g., an aspect), we propose an effective “perturb-and-see” method to extract sentiment words modifying it from large-scale datasets. The reliability of the obtained target-aware sentiment lexicons is extensively evaluated both manually and automatically. We also show that a simple application of the lexicon is able to achieve highly competitive performances on the unsupervised opinion relation extraction task.
Quality estimation aims to measure the quality of translated content without access to a reference translation. This is crucial for machine translation systems in real-world scenarios where high-quality translation is needed. While many approaches exist for quality estimation, they are based on supervised machine learning requiring costly human labelled data. As an alternative, we propose a technique that does not rely on examples from human-annotators and instead uses synthetic training data. We train off-the-shelf architectures for supervised quality estimation on our synthetic data and show that the resulting models achieve comparable performance to models trained on human-annotated data, both for sentence and word-level prediction.
This paper empirically studies whether BERT can really learn to conduct natural language inference (NLI) without utilizing hidden dataset bias; and how efficiently it can learn if it could. This is done via creating a simple entailment judgment case which involves only binary predicates in plain English. The results show that the learning process of BERT is very slow. However, the efficiency of learning can be greatly improved (data reduction by a factor of 1,500) if task-related features are added. This suggests that domain knowledge greatly helps when conducting NLI with neural networks.
We revisit the classic problem of document-level role-filler entity extraction (REE) for template filling. We argue that sentence-level approaches are ill-suited to the task and introduce a generative transformer-based encoder-decoder framework (GRIT) that is designed to model context at the document level: it can make extraction decisions across sentence boundaries; is implicitly aware of noun phrase coreference structure, and has the capacity to respect cross-role dependencies in the template structure. We evaluate our approach on the MUC-4 dataset, and show that our model performs substantially better than prior work. We also show that our modeling choices contribute to model performance, e.g., by implicitly capturing linguistic knowledge such as recognizing coreferent entity mentions.
Much research effort has been put to multilingual knowledge graph (KG) embedding methods to address the entity alignment task, which seeks to match entities in different languagespecific KGs that refer to the same real-world object. Such methods are often hindered by the insufficiency of seed alignment provided between KGs. Therefore, we propose a new model, JEANS , which jointly represents multilingual KGs and text corpora in a shared embedding scheme, and seeks to improve entity alignment with incidental supervision signals from text. JEANS first deploys an entity grounding process to combine each KG with the monolingual text corpus. Then, two learning processes are conducted: (i) an embedding learning process to encode the KG and text of each language in one embedding space, and (ii) a self-learning based alignment learning process to iteratively induce the correspondence of entities and that of lexemes between embeddings. Experiments on benchmark datasets show that JEANS leads to promising improvement on entity alignment with incidental supervision, and significantly outperforms state-of-the-art methods that solely rely on internal information of KGs.
This paper studies the problem of generatinglikely queries for multimodal documents withimages. Our application scenario is enablingefficient “first-stage retrieval” of relevant doc-uments, by attaching generated queries to doc-uments before indexing. We can then indexthis expanded text to efficiently narrow downto candidate matches using inverted index, sothat expensive reranking can follow. Our eval-uation results show that our proposed multi-modal representation meaningfully improvesrelevance ranking.More importantly, ourframework can achieve the state of the art inthe first stage retrieval scenarios
Non-neural approaches to argument mining (AM) are often pipelined and require heavy feature-engineering. In this paper, we propose a neural end-to-end approach to AM which is based on dependency parsing, in contrast to the current state-of-the-art which relies on relation extraction. Our biaffine AM dependency parser significantly outperforms the state-of-the-art, performing at F1 = 73.5% for component identification and F1 = 46.4% for relation identification. One of the advantages of treating AM as biaffine dependency parsing is the simple neural architecture that results. The idea of treating AM as dependency parsing is not new, but has previously been abandoned as it was lagging far behind the state-of-the-art. In a thorough analysis, we investigate the factors that contribute to the success of our model: the biaffine model itself, our representation for the dependency structure of arguments, different encoders in the biaffine model, and syntactic information additionally fed to the model. Our work demonstrates that dependency parsing for AM, an overlooked idea from the past, deserves more attention in the future.
Fake news articles often stir the readers’ attention by means of emotional appeals that arouse their feelings. Unlike in short news texts, authors of longer articles can exploit such affective factors to manipulate readers by adding exaggerations or fabricating events, in order to affect the readers’ emotions. To capture this, we propose in this paper to model the flow of affective information in fake news articles using a neural architecture. The proposed model, FakeFlow, learns this flow by combining topic and affective information extracted from text. We evaluate the model’s performance with several experiments on four real-world datasets. The results show that FakeFlow achieves superior results when compared against state-of-the-art methods, thus confirming the importance of capturing the flow of the affective information in news articles.
Previous studies demonstrated that a dynamic phone-informed compression of the input audio is beneficial for speech translation (ST). However, they required a dedicated model for phone recognition and did not test this solution for direct ST, in which a single model translates the input audio into the target language without intermediate representations. In this work, we propose the first method able to perform a dynamic compression of the input in direct ST models. In particular, we exploit the Connectionist Temporal Classification (CTC) to compress the input sequence according to its phonetic characteristics. Our experiments demonstrate that our solution brings a 1.3-1.5 BLEU improvement over a strong baseline on two language pairs (English-Italian and English-German), contextually reducing the memory footprint by more than 10%.
We present an open-source speech corpus for the Kazakh language. The Kazakh speech corpus (KSC) contains around 332 hours of transcribed audio comprising over 153,000 utterances spoken by participants from different regions and age groups, as well as both genders. It was carefully inspected by native Kazakh speakers to ensure high quality. The KSC is the largest publicly available database developed to advance various Kazakh speech and language processing applications. In this paper, we first describe the data collection and preprocessing procedures followed by a description of the database specifications. We also share our experience and challenges faced during the database construction, which might benefit other researchers planning to build a speech corpus for a low-resource language. To demonstrate the reliability of the database, we performed preliminary speech recognition experiments. The experimental results imply that the quality of audio and transcripts is promising (2.8% character error rate and 8.7% word error rate on the test set). To enable experiment reproducibility and ease the corpus usage, we also released an ESPnet recipe for our speech recognition models.
Tasks, Datasets and Evaluation Metrics are important concepts for understanding experimental scientific papers. However, previous work on information extraction for scientific literature mainly focuses on the abstracts only, and does not treat datasets as a separate type of entity (Zadeh and Schumann, 2016; Luan et al., 2018). In this paper, we present a new corpus that contains domain expert annotations for Task (T), Dataset (D), Metric (M) entities 2,000 sentences extracted from NLP papers. We report experiment results on TDM extraction using a simple data augmentation strategy and apply our tagger to around 30,000 NLP papers from the ACL Anthology. The corpus is made publicly available to the community for fostering research on scientific publication summarization (Erera et al., 2019) and knowledge discovery.
We introduce a top-down approach to discourse parsing that is conceptually simpler than its predecessors (Kobayashi et al., 2020; Zhang et al., 2020). By framing the task as a sequence labelling problem where the goal is to iteratively segment a document into individual discourse units, we are able to eliminate the decoder and reduce the search space for splitting points. We explore both traditional recurrent models and modern pre-trained transformer models for the task, and additionally introduce a novel dynamic oracle for top-down parsing. Based on the Full metric, our proposed LSTM model sets a new state-of-the-art for RST parsing.
Recent advancements in data-to-text generation largely take on the form of neural end-to-end systems. Efforts have been dedicated to improving text generation systems by changing the order of training samples in a process known as curriculum learning. Past research on sequence-to-sequence learning showed that curriculum learning helps to improve both the performance and convergence speed. In this work, we delve into the same idea surrounding the training samples consisting of structured data and text pairs, where at each update, the curriculum framework selects training samples based on the model’s competence. Specifically, we experiment with various difficulty metrics and put forward a soft edit distance metric for ranking training samples. On our benchmarks, it shows faster convergence speed where training time is reduced by 38.7% and performance is boosted by 4.84 BLEU.
We investigate how to solve the cross-corpus news recommendation for unseen users in the future. This is a problem where traditional content-based recommendation techniques often fail. Luckily, in real-world recommendation services, some publisher (e.g., Daily news) may have accumulated a large corpus with lots of consumers which can be used for a newly deployed publisher (e.g., Political news). To take advantage of the existing corpus, we propose a transfer learning model (dubbed as TrNews) for news recommendation to transfer the knowledge from a source corpus to a target corpus. To tackle the heterogeneity of different user interests and of different word distributions across corpora, we design a translator-based transfer-learning strategy to learn a representation mapping between source and target corpora. The learned translator can be used to generate representations for unseen users in the future. We show through experiments on real-world datasets that TrNews is better than various baselines in terms of four metrics. We also show that our translator is effective among existing transfer strategies.
Thanks to the success of goal-oriented negotiation dialogue systems, studies of negotiation dialogue have gained momentum in terms of both human-human negotiation support and dialogue systems. However, the field suffers from a paucity of available negotiation corpora, which hinders further development and makes it difficult to test new methodologies in novel negotiation settings. Here, we share a human-human negotiation dialogue dataset in a job interview scenario that features increased complexities in terms of the number of possible solutions and a utility function. We test the proposed corpus using a breakdown detection task for human-human negotiation support. We also introduce a dialogue act-based breakdown detection method, focusing on dialogue flow that is applicable to various corpora. Our results show that our proposed method features comparable detection performance to text-based approaches in existing corpora and better results in the proposed dataset.
For many new application domains for data-to-text generation, the main obstacle in training neural models consists of a lack of training data. While usually large numbers of instances are available on the data side, often only very few text samples are available. To address this problem, we here propose a novel few-shot approach for this setting. Our approach automatically augments the data available for training by (i) generating new text samples based on replacing specific values by alternative ones from the same category, (ii) generating new text samples based on GPT-2, and (iii) proposing an automatic method for pairing the new text samples with data samples. As the text augmentation can introduce noise to the training data, we use cycle consistency as an objective, in order to make sure that a given data sample can be correctly reconstructed after having been formulated as text (and that text samples can be reconstructed from data). On both the E2E and WebNLG benchmarks, we show that this weakly supervised training paradigm is able to outperform fully supervised sequence-to-sequence models with less than 10% of the training set. By utilizing all annotated data, our model can boost the performance of a standard sequence-to-sequence model by over 5 BLEU points, establishing a new state-of-the-art on both datasets.
A sufficient amount of annotated data is usually required to fine-tune pre-trained language models for downstream tasks. Unfortunately, attaining labeled data can be costly, especially for multiple language varieties and dialects. We propose to self-train pre-trained language models in zero- and few-shot scenarios to improve performance on data-scarce varieties using only resources from data-rich ones. We demonstrate the utility of our approach in the context of Arabic sequence labeling by using a language model fine-tuned on Modern Standard Arabic (MSA) only to predict named entities (NE) and part-of-speech (POS) tags on several dialectal Arabic (DA) varieties. We show that self-training is indeed powerful, improving zero-shot MSA-to-DA transfer by as large as ˷10% F1 (NER) and 2% accuracy (POS tagging). We acquire even better performance in few-shot scenarios with limited amounts of labeled data. We conduct an ablation study and show that the performance boost observed directly results from training data augmentation possible with DA examples via self-training. This opens up opportunities for developing DA models exploiting only MSA resources. Our approach can also be extended to other languages and tasks.
Departing from both sequential pipelines and monotask systems, we propose Multiple Tasks Integration (MTI), a multitask paradigm orthogonal to weight sharing. The essence of MTI is to process the input iteratively but concurrently at multiple levels of analysis, where each decision is based on all of the structures that are already inferred and free from usual ordering constraints. We illustrate MTI with a system that performs part-of-speech tagging, syntactic dependency parsing and semantic dependency parsing. We observe that both the use of reinforcement learning and the release from sequential constraints are beneficial to the quality of the syntactic and semantic parses. We also observe that our model adopts an easy-first strategy that consists, on average, of predicting shorter dependencies before longer ones, but that syntax is not always tackled before semantics.
In this paper, we address the representation of coordinate constructions in Enhanced Universal Dependencies (UD), where relevant dependency links are propagated from conjunction heads to other conjuncts. English treebanks for enhanced UD have been created from gold basic dependencies using a heuristic rule-based converter, which propagates only core arguments. With the aim of determining which set of links should be propagated from a semantic perspective, we create a large-scale dataset of manually edited syntax graphs. We identify several systematic errors in the original data, and propose to also propagate adjuncts. We observe high inter-annotator agreement for this semantic annotation task. Using our new manually verified dataset, we perform the first principled comparison of rule-based and (partially novel) machine-learning based methods for conjunction propagation for English. We show that learning propagation rules is more effective than hand-designing heuristic rules. When using automatic parses, our neural graph-parser based edge predictor outperforms the currently predominant pipelines using a basic-layer tree parser plus converters.
Most, if not all forms of ellipsis (e.g., so does Mary) are similar to reading comprehension questions (what does Mary do), in that in order to resolve them, we need to identify an appropriate text span in the preceding discourse. Following this observation, we present an alternative approach for English ellipsis resolution relying on architectures developed for question answering (QA). We present both single-task models, and joint models trained on auxiliary QA and coreference resolution datasets, clearly outperforming the current state of the art for Sluice Ellipsis (from 70.00 to 86.01 F1) and Verb Phrase Ellipsis (from 72.89 to 78.66 F1).
Neural natural language generation (NLG) and understanding (NLU) models are data-hungry and require massive amounts of annotated data to be competitive. Recent frameworks address this bottleneck with generative models that synthesize weak labels at scale, where a small amount of training labels are expert-curated and the rest of the data is automatically annotated. We follow that approach, by automatically constructing a large-scale weakly-labeled data with a fine-tuned GPT-2, and employ a semi-supervised framework to jointly train the NLG and NLU models. The proposed framework adapts the parameter updates to the models according to the estimated label-quality. On both the E2E and Weather benchmarks, we show that this weakly supervised training paradigm is an effective approach under low resource scenarios with as little as 10 data instances, and outperforming benchmark systems on both datasets when 100% of the training data is used.
While recent advances in deep learning led to significant improvements in machine translation, neural machine translation is often still not able to continuously adapt to the environment. For humans, as well as for machine translation, bilingual dictionaries are a promising knowledge source to continuously integrate new knowledge. However, their exploitation poses several challenges: The system needs to be able to perform one-shot learning as well as model the morphology of source and target language. In this work, we proposed an evaluation framework to assess the ability of neural machine translation to continuously learn new phrases. We integrate one-shot learning methods for neural machine translation with different word representations and show that it is important to address both in order to successfully make use of bilingual dictionaries. By addressing both challenges we are able to improve the ability to translate new, rare words and phrases from 30% to up to 70%. The correct lemma is even generated by more than 90%.
Deep learning models are susceptible to adversarial examples that have imperceptible perturbations in the original input, resulting in adversarial attacks against these models. Analysis of these attacks on the state of the art transformers in NLP can help improve the robustness of these models against such adversarial inputs. In this paper, we present Adv-OLM, a black-box attack method that adapts the idea of Occlusion and Language Models (OLM) to the current state of the art attack methods. OLM is used to rank words of a sentence, which are later substituted using word replacement strategies. We experimentally show that our approach outperforms other attack methods for several text classification tasks.
This paper addresses the task of (complex) conversational question answering over a knowledge graph. For this task, we propose LASAGNE (muLti-task semAntic parSing with trAnsformer and Graph atteNtion nEworks). It is the first approach, which employs a transformer architecture extended with Graph Attention Networks for multi-task neural semantic parsing. LASAGNE uses a transformer model for generating the base logical forms, while the Graph Attention model is used to exploit correlations between (entity) types and predicates to produce node representations. LASAGNE also includes a novel entity recognition module which detects, links, and ranks all relevant entities in the question context. We evaluate LASAGNE on a standard dataset for complex sequential question answering, on which it outperforms existing baselines averaged on all question types. Specifically, we show that LASAGNE improves the F1-score on eight out of ten question types; in some cases, the increase is more than 20% compared to state of the art (SotA).
Author stylized rewriting is the task of rewriting an input text in a particular author’s style. Recent works in this area have leveraged Transformer-based language models in a denoising autoencoder setup to generate author stylized text without relying on a parallel corpus of data. However, these approaches are limited by the lack of explicit control of target attributes and being entirely data-driven. In this paper, we propose a Director-Generator framework to rewrite content in the target author’s style, specifically focusing on certain target attributes. We show that our proposed framework works well even with a limited-sized target author corpus. Our experiments on corpora consisting of relatively small-sized text authored by three distinct authors show significant improvements upon existing works to rewrite input texts in target author’s style. Our quantitative and qualitative analyses further show that our model has better meaning retention and results in more fluent generations.
Generative models for open domain question answering have proven to be competitive, without resorting to external knowledge. While promising, this approach requires to use models with billions of parameters, which are expensive to train and query. In this paper, we investigate how much these models can benefit from retrieving text passages, potentially containing evidence. We obtain state-of-the-art results on the Natural Questions and TriviaQA open benchmarks. Interestingly, we observe that the performance of this method significantly improves when increasing the number of retrieved passages. This is evidence that sequence-to-sequence models offers a flexible framework to efficiently aggregate and combine evidence from multiple passages.
Outcome prediction from clinical text can prevent doctors from overlooking possible risks and help hospitals to plan capacities. We simulate patients at admission time, when decision support can be especially valuable, and contribute a novel *admission to discharge* task with four common outcome prediction targets: Diagnoses at discharge, procedures performed, in-hospital mortality and length-of-stay prediction. The ideal system should infer outcomes based on symptoms, pre-conditions and risk factors of a patient. We evaluate the effectiveness of language models to handle this scenario and propose *clinical outcome pre-training* to integrate knowledge about patient outcomes from multiple public sources. We further present a simple method to incorporate ICD code hierarchy into the models. We show that our approach improves performance on the outcome tasks against several baselines. A detailed analysis reveals further strengths of the model, including transferability, but also weaknesses such as handling of vital values and inconsistencies in the underlying data.
Semi-supervised learning through deep generative models and multi-lingual pretraining techniques have orchestrated tremendous success across different areas of NLP. Nonetheless, their development has happened in isolation, while the combination of both could potentially be effective for tackling task-specific labelled data shortage. To bridge this gap, we combine semi-supervised deep generative models and multi-lingual pretraining to form a pipeline for document classification task. Compared to strong supervised learning baselines, our semi-supervised classification framework is highly competitive and outperforms the state-of-the-art counterparts in low-resource settings across several languages.
Universal schema (USchema) assumes that two sentence patterns that share the same entity pairs are similar to each other. This assumption is widely adopted for solving various types of relation extraction (RE) tasks. Nevertheless, each sentence pattern could contain multiple facets, and not every facet is similar to all the facets of another sentence pattern co-occurring with the same entity pair. To address the violation of the USchema assumption, we propose multi-facet universal schema that uses a neural model to represent each sentence pattern as multiple facet embeddings and encourage one of these facet embeddings to be close to that of another sentence pattern if they co-occur with the same entity pair. In our experiments, we demonstrate that multi-facet embeddings significantly outperform their single-facet embedding counterpart, compositional universal schema (CUSchema) (Verga et al., 2016), in distantly supervised relation extraction tasks. Moreover, we can also use multiple embeddings to detect the entailment relation between two sentence patterns when no manual label is available.
Despite the recent success of deep neural networks in natural language processing, the extent to which they can demonstrate human-like generalization capacities for natural language understanding remains unclear. We explore this issue in the domain of natural language inference (NLI), focusing on the transitivity of inference relations, a fundamental property for systematically drawing inferences. A model capturing transitivity can compose basic inference patterns and draw new inferences. We introduce an analysis method using synthetic and naturalistic NLI datasets involving clause-embedding verbs to evaluate whether models can perform transitivity inferences composed of veridical inferences and arbitrary inference types. We find that current NLI models do not perform consistently well on transitivity inference tasks, suggesting that they lack the generalization capacity for drawing composite inferences from provided training examples. The data and code for our analysis are publicly available at https://github.com/verypluming/transitivity.
Modern classification models tend to struggle when the amount of annotated data is scarce. To overcome this issue, several neural few-shot classification models have emerged, yielding significant progress over time, both in Computer Vision and Natural Language Processing. In the latter, such models used to rely on fixed word embeddings, before the advent of transformers. Additionally, some models used in Computer Vision are yet to be tested in NLP applications. In this paper, we compare all these models, first adapting those made in the field of image processing to NLP, and second providing them access to transformers. We then test these models equipped with the same transformer-based encoder on the intent detection task, known for having a large amount of classes. Our results reveal that while methods perform almost equally on the ARSC dataset, this is not the case for the Intent Detection task, where most recent and supposedly best competitors perform worse than older and simpler ones (while all are are given access to transformers). We also show that a simple baseline is surprisingly strong. All the new developed models as well as the evaluation framework are made publicly available.
State-of-the-art multilingual machine translation relies on a universal encoder-decoder, which requires retraining the entire system to add new languages. In this paper, we propose an alternative approach that is based on language-specific encoder-decoders, and can thus be more easily extended to new languages by learning their corresponding modules. So as to encourage a common interlingua representation, we simultaneously train the N initial languages. Our experiments show that the proposed approach outperforms the universal encoder-decoder by 3.28 BLEU points on average, while allowing to add new languages without the need to retrain the rest of the modules. All in all, our work closes the gap between shared and language-specific encoderdecoders, advancing toward modular multilingual machine translation systems that can be flexibly extended in lifelong learning settings.
Romanian is one of the understudied languages in computational linguistics, with few resources available for the development of natural language processing tools. In this paper, we introduce LaRoSeDa, a Large Romanian Sentiment Data Set, which is composed of 15,000 positive and negative reviews collected from the largest Romanian e-commerce platform. We employ two sentiment classification methods as baselines for our new data set, one based on low-level features (character n-grams) and one based on high-level features (bag-of-word-embeddings generated by clustering word embeddings with k-means). As an additional contribution, we replace the k-means clustering algorithm with self-organizing maps (SOMs), obtaining better results because the generated clusters of word embeddings are closer to the Zipf’s law distribution, which is known to govern natural language. We also demonstrate the generalization capacity of using SOMs for the clustering of word embeddings on another recently-introduced Romanian data set, for text categorization by topic.
The biases present in training datasets have been shown to affect models for sentence pair classification tasks such as natural language inference (NLI) and fact verification. While fine-tuning models on additional data has been used to mitigate them, a common issue is that of catastrophic forgetting of the original training dataset. In this paper, we show that elastic weight consolidation (EWC) allows fine-tuning of models to mitigate biases while being less susceptible to catastrophic forgetting. In our evaluation on fact verification and NLI stress tests, we show that fine-tuning with EWC dominates standard fine-tuning, yielding models with lower levels of forgetting on the original (biased) dataset for equivalent gains in accuracy on the fine-tuning (unbiased) dataset.
The widespread of fake news and misinformation in various domains ranging from politics, economics to public health has posed an urgent need to automatically fact-check information. A recent trend in fake news detection is to utilize evidence from external sources. However, existing evidence-aware fake news detection methods focused on either only word-level attention or evidence-level attention, which may result in suboptimal performance. In this paper, we propose a Hierarchical Multi-head Attentive Network to fact-check textual claims. Our model jointly combines multi-head word-level attention and multi-head document-level attention, which aid explanation in both word-level and evidence-level. Experiments on two real-word datasets show that our model outperforms seven state-of-the-art baselines. Improvements over baselines are from 6% to 18%. Our source code and datasets are released at https://github.com/nguyenvo09/EACL2021.
Identifying named entities in written text is an essential component of the text processing pipeline used in applications such as text editors to gain a better understanding of the semantics of the text. However, the typical experimental setup for evaluating Named Entity Recognition (NER) systems is not directly applicable to systems that process text in real time as the text is being typed. Evaluation is performed on a sentence level assuming the end-user is willing to wait until the entire sentence is typed for entities to be identified and further linked to identifiers or co-referenced. We introduce a novel experimental setup for NER systems for applications where decisions about named entity boundaries need to be performed in an online fashion. We study how state-of-the-art methods perform under this setup in multiple languages and propose adaptations to these models to suit this new experimental setup. Experimental results show that the best systems that are evaluated on each token after its typed, reach performance within 1–5 F1 points of systems that are evaluated at the end of the sentence. These show that entity recognition can be performed in this setup and open up the development of other NLP tools in a similar setup.
The Internet contains a multitude of social media posts and other of stories where text is interspersed with images. In these contexts, images are not simply used for general illustration, but are judiciously placed in certain spots of a story for multimodal descriptions and narration. In this work we analyze the problem of text-image alignment, and present SANDI, a methodology for automatically selecting images from an image collection and aligning them with text paragraphs of a story. SANDI combines visual tags, user-provided tags and background knowledge, and uses an Integer Linear Program to compute alignments that are semantically meaningful. Experiments show that SANDI can select and align images with texts with high quality of semantic fit.
Ideally Open-Domain Question Answering models should exhibit a number of competencies, ranging from simply memorizing questions seen at training time, to answering novel question formulations with answers seen during training, to generalizing to completely novel questions with novel answers. However, single aggregated test set scores do not show the full picture of what capabilities models truly have. In this work, we perform a detailed study of the test sets of three popular open-domain benchmark datasets with respect to these competencies. We find that 30% of test-set questions have a near-duplicate paraphrase in their corresponding train sets. In addition, we find that 60-70% of answers in the test sets are also present in the train sets. Using these findings, we evaluate a variety of popular open-domain models to obtain greater insight into what extent they can generalize, and what drives their overall performance. We find that all models perform substantially worse on questions that cannot be memorized from train sets, with a mean absolute performance difference of 61% between repeated and non-repeated data. Finally we show that simple nearest-neighbor models outperform a BART closed-book QA model, further highlighting the role that train set memorization plays in these benchmarks
Being able to parse code-switched (CS) utterances, such as Spanish+English or Hindi+English, is essential to democratize task-oriented semantic parsing systems for certain locales. In this work, we focus on Spanglish (Spanish+English) and release a dataset, CSTOP, containing 5800 CS utterances alongside their semantic parses. We examine the CS generalizability of various Cross-lingual (XL) models and exhibit the advantage of pre-trained XL language models when data for only one language is present. As such, we focus on improving the pre-trained models for the case when only English corpus alongside either zero or a few CS training instances are available. We propose two data augmentation methods for the zero-shot and the few-shot settings: fine-tune using translate-and-align and augment using a generation model followed by match-and-filter. Combining the few-shot setting with the above improvements decreases the initial 30-point accuracy gap between the zero-shot and the full-data settings by two thirds.
Paraphrase generation plays an essential role in natural language process (NLP), and it has many downstream applications. However, training supervised paraphrase models requires many annotated paraphrase pairs, which are usually costly to obtain. On the other hand, the paraphrases generated by existing unsupervised approaches are usually syntactically similar to the source sentences and are limited in diversity. In this paper, we demonstrate that it is possible to generate syntactically various paraphrases without the need for annotated paraphrase pairs. We propose Syntactically controlled Paraphrase Generator (SynPG), an encoder-decoder based model that learns to disentangle the semantics and the syntax of a sentence from a collection of unannotated texts. The disentanglement enables SynPG to control the syntax of output paraphrases by manipulating the embedding in the syntactic space. Extensive experiments using automatic metrics and human evaluation show that SynPG performs better syntactic control than unsupervised baselines, while the quality of the generated paraphrases is competitive. We also demonstrate that the performance of SynPG is competitive or even better than supervised models when the unannotated data is large. Finally, we show that the syntactically controlled paraphrases generated by SynPG can be utilized for data augmentation to improve the robustness of NLP models.
The automatic detection of hypernymy relationships represents a challenging problem in NLP. The successful application of state-of-the-art supervised approaches using distributed representations has generally been impeded by the limited availability of high quality training data. We have developed two novel data augmentation techniques which generate new training examples from existing ones. First, we combine the linguistic principles of hypernym transitivity and intersective modifier-noun composition to generate additional pairs of vectors, such as “small dog - dog” or “small dog - animal”, for which a hypernymy relationship can be assumed. Second, we use generative adversarial networks (GANs) to generate pairs of vectors for which the hypernymy relation can also be assumed. We furthermore present two complementary strategies for extending an existing dataset by leveraging linguistic resources such as WordNet. Using an evaluation across 3 different datasets for hypernymy detection and 2 different vector spaces, we demonstrate that both of the proposed automatic data augmentation and dataset extension strategies substantially improve classifier performance.
Machine translation (MT) models used in industries with constantly changing topics, such as translation or news agencies, need to adapt to new data to maintain their performance over time. Our aim is to teach a pre-trained MT model to translate previously unseen words accurately, based on very few examples. We propose (i) an experimental setup allowing us to simulate novel vocabulary appearing in human-submitted translations, and (ii) corresponding evaluation metrics to compare our approaches. We extend a data augmentation approach using a pretrained language model to create training examples with similar contexts for novel words. We compare different fine-tuning and data augmentation approaches and show that adaptation on the scale of one to five examples is possible. Combining data augmentation with randomly selected training sentences leads to the highest BLEU score and accuracy improvements. Impressively, with only 1 to 5 examples, our model reports better accuracy scores than a reference system trained with on average 313 parallel examples.
Dialog State Tracking (DST), an integral part of modern dialog systems, aims to track user preferences and constraints (slots) in task-oriented dialogs. In real-world settings with constantly changing services, DST systems must generalize to new domains and unseen slot types. Existing methods for DST do not generalize well to new slot names and many require known ontologies of slot types and values for inference. We introduce a novel ontology-free framework that supports natural language queries for unseen constraints and slots in multi-domain task-oriented dialogs. Our approach is based on generative question-answering using a conditional language model pre-trained on substantive English sentences. Our model improves joint goal accuracy in zero-shot domain adaptation settings by up to 9% (absolute) over the previous state-of-the-art on the MultiWOZ 2.1 dataset.
A major obstacle to the wide-spread adoption of neural retrieval models is that they require large supervised training sets to surpass traditional term-based techniques, which are constructed from raw corpora. In this paper, we propose an approach to zero-shot learning for passage retrieval that uses synthetic question generation to close this gap. The question generation system is trained on general domain data, but is applied to documents in the targeted domain. This allows us to create arbitrarily large, yet noisy, question-passage relevance pairs that are domain specific. Furthermore, when this is coupled with a simple hybrid term-neural model, first-stage retrieval performance can be improved further. Empirically, we show that this is an effective strategy for building neural passage retrieval models in the absence of large training corpora. Depending on the domain, this technique can even approach the accuracy of supervised models.
We propose an unsupervised graph-based ranking model for extractive summarization of long scientific documents. Our method assumes a two-level hierarchical graph representation of the source document, and exploits asymmetrical positional cues to determine sentence importance. Results on the PubMed and arXiv datasets show that our approach outperforms strong unsupervised baselines by wide margins in automatic metrics and human evaluation. In addition, it achieves performance comparable to many state-of-the-art supervised approaches which are trained on hundreds of thousands of examples. These results suggest that patterns in the discourse structure are a strong signal for determining importance in scientific articles.
Dialog act prediction in open-domain conversations is an essential language comprehension task for both dialog system building and discourse analysis. Previous dialog act schemes, such as SWBD-DAMSL, are designed mainly for discourse analysis in human-human conversations. In this paper, we present a dialog act annotation scheme, MIDAS (Machine Interaction Dialog Act Scheme), targeted at open-domain human-machine conversations. MIDAS is designed to assist machines to improve their ability to understand human partners. MIDAS has a hierarchical structure and supports multi-label annotations. We collected and annotated a large open-domain human-machine spoken conversation dataset (consisting of 24K utterances). To validate our scheme, we leveraged transfer learning methods to train a multi-label dialog act prediction model and reached an F1 score of 0.79.
In this work, we study how the finetuning stage in the pretrain-finetune framework changes the behavior of a pretrained neural language generator. We focus on the transformer encoder-decoder model for the open-domain dialogue response generation task. Our major finding is that after standard finetuning, the model forgets some of the important language generation skills acquired during large-scale pretraining. We demonstrate the forgetting phenomenon through a set of detailed behavior analysis from the perspectives of knowledge transfer, context sensitivity, and function space projection. As a preliminary attempt to alleviate the forgetting problem, we propose an intuitive finetuning strategy named “mix-review”. We find that mix-review effectively regularizes the finetuning process, and the forgetting problem is alleviated to some extent. Finally, we discuss interesting behavior of the resulting dialogue model and its implications.
“Transcription bottlenecks”, created by a shortage of effective human transcribers (i.e., transcriber shortage), are one of the main challenges to endangered language (EL) documentation. Automatic speech recognition (ASR) has been suggested as a tool to overcome such bottlenecks. Following this suggestion, we investigated the effectiveness for EL documentation of end-to-end ASR, which unlike Hidden Markov Model ASR systems, eschews linguistic resources but is instead more dependent on large-data settings. We open source a Yoloxóchitl Mixtec EL corpus. First, we review our method in building an end-to-end ASR system in a way that would be reproducible by the ASR community. We then propose a novice transcription correction task and demonstrate how ASR systems and novice transcribers can work together to improve EL documentation. We believe this combinatory methodology would mitigate the transcription bottleneck and transcriber shortage that hinders EL documentation.
The success of authorship attribution relies on the presence of linguistic features specific to individual authors. There is, however, limited research assessing to what extent authorial style remains constant when individuals switch from one writing modality to another. We measure the effect of writing mode on writing style in the context of authorship attribution research using a corpus of documents composed online (in a web browser) and documents composed offline using a traditional word processor. The results confirm the existence of a “mode effect” on authorial style. Online writing differs systematically from offline writing in terms of sentence length, word use, readability, and certain part-of-speech ratios. These findings have implications for research design and feature engineering in authorship attribution studies.
Latent variable models for text, when trained successfully, accurately model the data distribution and capture global semantic and syntactic features of sentences. The prominent approach to train such models is variational autoencoders (VAE). It is nevertheless challenging to train and often results in a trivial local optimum where the latent variable is ignored and its posterior collapses into the prior, an issue known as posterior collapse. Various techniques have been proposed to mitigate this issue. Most of them focus on improving the inference model to yield latent codes of higher quality. The present work proposes a short run dynamics for inference. It is initialized from the prior distribution of the latent variable and then runs a small number (e.g., 20) of Langevin dynamics steps guided by its posterior distribution. The major advantage of our method is that it does not require a separate inference model or assume simple geometry of the posterior distribution, thus rendering an automatic, natural and flexible inference engine. We show that the models trained with short run dynamics more accurately model the data, compared to strong language model and VAE baselines, and exhibit no sign of posterior collapse. Analyses of the latent space show that interpolation in the latent space is able to generate coherent sentences with smooth transition and demonstrate improved classification over strong baselines with latent features from unsupervised pretraining. These results together expose a well-structured latent space of our generative model.
Podcast episodes often contain material extraneous to the main content, such as advertisements, interleaved within the audio and the written descriptions. We present classifiers that leverage both textual and listening patterns in order to detect such content in podcast descriptions and audio transcripts. We demonstrate that our models are effective by evaluating them on the downstream task of podcast summarization and show that we can substantively improve ROUGE scores and reduce the extraneous content generated in the summaries.
Expressive text encoders such as RNNs and Transformer Networks have been at the center of NLP models in recent work. Most of the effort has focused on sentence-level tasks, capturing the dependencies between words in a single sentence, or pairs of sentences. However, certain tasks, such as argumentation mining, require accounting for longer texts and complicated structural dependencies between them. Deep structured prediction is a general framework to combine the complementary strengths of expressive neural encoders and structured inference for highly structured domains. Nevertheless, when the need arises to go beyond sentences, most work relies on combining the output scores of independently trained classifiers. One of the main reasons for this is that constrained inference comes at a high computational cost. In this paper, we explore the use of randomized inference to alleviate this concern and show that we can efficiently leverage deep structured prediction and expressive neural encoders for a set of tasks involving complicated argumentative structures.
We propose to leverage lexical paraphrases and high precision rules informed by news discourse structure to automatically collect coreferential and non-coreferential event pairs from unlabeled English news articles. We perform both manual validation and empirical evaluation on multiple evaluation datasets with different event domains and text genres to assess the quality of our acquired event pairs. We found that a model trained on our acquired event pairs performs comparably as the supervised model when applied to new data out of the training data domains. Further, augmenting human-annotated data with the acquired event pairs provides empirical performance gains on both in-domain and out-of-domain evaluation datasets.
Existing approaches for table annotation with entities and types either capture the structure of table using graphical models, or learn embeddings of table entries without accounting for the complete syntactic structure. We propose TabGCN, that uses Graph Convolutional Networks to capture the complete structure of tables, knowledge graph and the training annotations, and jointly learns embeddings for table elements as well as the entities and types. To account for knowledge incompleteness, TabGCN’s embeddings can be used to discover new entities and types. Using experiments on 5 benchmark datasets, we show that TabGCN significantly outperforms multiple state-of-the-art baselines for table annotation, while showing promising performance on downstream table-related applications.
One of the most challenging topics in Natural Language Processing (NLP) is visually-grounded language understanding and reasoning. Outdoor vision-and-language navigation (VLN) is such a task where an agent follows natural language instructions and navigates in real-life urban environments. With the lack of human-annotated instructions that illustrate the intricate urban scenes, outdoor VLN remains a challenging task to solve. In this paper, we introduce a Multimodal Text Style Transfer (MTST) learning approach and leverage external multimodal resources to mitigate data scarcity in outdoor navigation tasks. We first enrich the navigation data by transferring the style of the instructions generated by Google Maps API, then pre-train the navigator with the augmented external outdoor navigation dataset. Experimental results show that our MTST learning approach is model-agnostic, and our MTST approach significantly outperforms the baseline models on the outdoor VLN task, improving task completion rate by 8.7% relatively on the test set.
Novel Object Captioning is a zero-shot Image Captioning task requiring describing objects not seen in the training captions, but for which information is available from external object detectors. The key challenge is to select and describe all salient detected novel objects in the input images. In this paper, we focus on this challenge and propose the ECOL-R model (Encouraging Copying of Object Labels with Reinforced Learning), a copy-augmented transformer model that is encouraged to accurately describe the novel object labels. This is achieved via a specialised reward function in the SCST reinforcement learning framework (Rennie et al., 2017) that encourages novel object mentions while maintaining the caption quality. We further restrict the SCST training to the images where detected objects are mentioned in reference captions to train the ECOL-R model. We additionally improve our copy mechanism via Abstract Labels, which transfer knowledge from known to novel object types, and a Morphological Selector, which determines the appropriate inflected forms of novel object labels. The resulting model sets new state-of-the-art on the nocaps (Agrawal et al., 2019) and held-out COCO (Hendricks et al., 2016) benchmarks.
The non-autoregressive models have boosted the efficiency of neural machine translation through parallelized decoding at the cost of effectiveness, when comparing with the autoregressive counterparts. In this paper, we claim that the syntactic and semantic structures among natural language are critical for non-autoregressive machine translation and can further improve the performance. However, these structures are rarely considered in the existing non-autoregressive models. Inspired by this intuition, we propose to incorporate the explicit syntactic and semantic structure of languages into a non-autoregressive Transformer, for the task of neural machine translation. Moreover, we also consider the intermediate latent alignment within target sentences to better learn the long-term token dependencies. Experimental results on two real-world datasets (i.e., WMT14 En-De and WMT16 En- Ro) show that our model achieves a significantly faster speed, as well as keeps the translation quality when compared with several state-of-the-art non-autoregressive models.
We introduce NLQuAD, the first data set with baseline methods for non-factoid long question answering, a task requiring document-level language understanding. In contrast to existing span detection question answering data sets, NLQuAD has non-factoid questions that are not answerable by a short span of text and demanding multiple-sentence descriptive answers and opinions. We show the limitation of the F1 score for evaluation of long answers and introduce Intersection over Union (IoU), which measures position-sensitive overlap between the predicted and the target answer spans. To establish baseline performances, we compare BERT, RoBERTa, and Longformer models. Experimental results and human evaluations show that Longformer outperforms the other architectures, but results are still far behind a human upper bound, leaving substantial room for improvements. NLQuAD’s samples exceed the input limitation of most pre-trained Transformer-based models, encouraging future research on long sequence language models.
In comparison to the numerous debiasing methods proposed for the static non-contextualised word embeddings, the discriminative biases in contextualised embeddings have received relatively little attention. We propose a fine-tuning method that can be applied at token- or sentence-levels to debias pre-trained contextualised embeddings. Our proposed method can be applied to any pre-trained contextualised embedding model, without requiring to retrain those models. Using gender bias as an illustrative example, we then conduct a systematic study using several state-of-the-art (SoTA) contextualised representations on multiple benchmark datasets to evaluate the level of biases encoded in different contextualised embeddings before and after debiasing using the proposed method. We find that applying token-level debiasing for all tokens and across all layers of a contextualised embedding model produces the best performance. Interestingly, we observe that there is a trade-off between creating an accurate vs. unbiased contextualised embedding model, and different contextualised embedding models respond differently to this trade-off.
Lexical inference in context (LIiC) is the task of recognizing textual entailment between two very similar sentences, i.e., sentences that only differ in one expression. It can therefore be seen as a variant of the natural language inference task that is focused on lexical semantics. We formulate and evaluate the first approaches based on pretrained language models (LMs) for this task: (i) a few-shot NLI classifier, (ii) a relation induction approach based on handcrafted patterns expressing the semantics of lexical inference, and (iii) a variant of (ii) with patterns that were automatically extracted from a corpus. All our approaches outperform the previous state of the art, showing the potential of pretrained LMs for LIiC. In an extensive analysis, we investigate factors of success and failure of our three approaches.
In this work, we investigate the problems of semantic parsing in a few-shot learning setting. In this setting, we are provided with k utterance-logical form pairs per new predicate. The state-of-the-art neural semantic parsers achieve less than 25% accuracy on benchmark datasets when k = 1. To tackle this problem, we proposed to i) apply a designated meta-learning method to train the model; ii) regularize attention scores with alignment statistics; iii) apply a smoothing technique in pretraining. As a result, our method consistently outperforms all the baselines in both one and two-shot settings.
Existing dialog system models require extensive human annotations and are difficult to generalize to different tasks. The recent success of large pre-trained language models such as BERT and GPT-2 (Devlin et al., 2019; Radford et al., 2019) have suggested the effectiveness of incorporating language priors in down-stream NLP tasks. However, how much pre-trained language models can help dialog response generation is still under exploration. In this paper, we propose a simple, general, and effective framework: Alternating Recurrent Dialog Model (ARDM). ARDM models each speaker separately and takes advantage of the large pre-trained language model. It requires no supervision from human annotations such as belief states or dialog acts to achieve effective conversations. ARDM outperforms or is on par with state-of-the-art methods on two popular task-oriented dialog datasets: CamRest676 and MultiWOZ. Moreover, we can generalize ARDM to more challenging, non-collaborative tasks such as persuasion. In persuasion tasks, ARDM is capable of generating human-like responses to persuade people to donate to a charity.
Vision-and-Language Navigation wayfinding agents can be enhanced by exploiting automatically generated navigation instructions. However, existing instruction generators have not been comprehensively evaluated, and the automatic evaluation metrics used to develop them have not been validated. Using human wayfinders, we show that these generators perform on par with or only slightly better than a template-based generator and far worse than human instructors. Furthermore, we discover that BLEU, ROUGE, METEOR and CIDEr are ineffective for evaluating grounded navigation instructions. To improve instruction evaluation, we propose an instruction-trajectory compatibility model that operates without reference instructions. Our model shows the highest correlation with human wayfinding outcomes when scoring individual instructions. For ranking instruction generation systems, if reference instructions are available we recommend using SPICE.
Pre-trained language models have been shown to improve performance in many natural language tasks substantially. Although the early focus of such models was single language pre-training, recent advances have resulted in cross-lingual and visual pre-training methods. In this paper, we combine these two approaches to learn visually-grounded cross-lingual representations. Specifically, we extend the translation language modelling (Lample and Conneau, 2019) with masked region classification and perform pre-training with three-way parallel vision & language corpora. We show that when fine-tuned for multimodal machine translation, these models obtain state-of-the-art performance. We also provide qualitative insights into the usefulness of the learned grounded representations.
Public datasets are often used to evaluate the efficacy and generalizability of state-of-the-art methods for many tasks in natural language processing (NLP). However, the presence of overlap between the train and test datasets can lead to inflated results, inadvertently evaluating the model’s ability to memorize and interpreting it as the ability to generalize. In addition, such data sets may not provide an effective indicator of the performance of these methods in real world scenarios. We identify leakage of training data into test data on several publicly available datasets used to evaluate NLP tasks, including named entity recognition and relation extraction, and study them to assess the impact of that leakage on the model’s ability to memorize versus generalize.
Online misogyny is a pernicious social problem that risks making online platforms toxic and unwelcoming to women. We present a new hierarchical taxonomy for online misogyny, as well as an expert labelled dataset to enable automatic classification of misogynistic content. The dataset consists of 6567 labels for Reddit posts and comments. As previous research has found untrained crowdsourced annotators struggle with identifying misogyny, we hired and trained annotators and provided them with robust annotation guidelines. We report baseline classification performance on the binary classification task, achieving accuracy of 0.93 and F1 of 0.43. The codebook and datasets are made freely available for future researchers.
We present an approach based on multilingual sentence embeddings to automatically extract parallel sentences from the content of Wikipedia articles in 96 languages, including several dialects or low-resource languages. We do not limit the extraction process to alignments with English, but we systematically consider all possible language pairs. In total, we are able to extract 135M parallel sentences for 16720 different language pairs, out of which only 34M are aligned with English. This corpus is freely available. To get an indication on the quality of the extracted bitexts, we train neural MT baseline systems on the mined data only for 1886 languages pairs, and evaluate them on the TED corpus, achieving strong BLEU scores for many language pairs. The WikiMatrix bitexts seem to be particularly interesting to train MT systems between distant languages without the need to pivot through English.
Chemical patents contain rich coreference and bridging links, which are the target of this research. Specially, we introduce a novel annotation scheme, based on which we create the ChEMU-Ref dataset from reaction description snippets in English-language chemical patents. We propose a neural approach to anaphora resolution, which we show to achieve strong results, especially when jointly trained over coreference and bridging links.
Standard models for syntactic dependency parsing take words to be the elementary units that enter into dependency relations. In this paper, we investigate whether there are any benefits from enriching these models with the more abstract notion of nucleus proposed by Tesnière. We do this by showing how the concept of nucleus can be defined in the framework of Universal Dependencies and how we can use composition functions to make a transition-based dependency parser aware of this concept. Experiments on 12 languages show that nucleus composition gives small but significant improvements in parsing accuracy. Further analysis reveals that the improvement mainly concerns a small number of dependency relations, including nominal modifiers, relations of coordination, main predicates, and direct objects.
Neural sequence-to-sequence models are currently the predominant choice for language generation tasks. Yet, on word-level tasks, exact inference of these models reveals the empty string is often the global optimum. Prior works have speculated this phenomenon is a result of the inadequacy of neural models for language generation. However, in the case of morphological inflection, we find that the empty string is almost never the most probable solution under the model. Further, greedy search often finds the global optimum. These observations suggest that the poor calibration of many neural models may stem from characteristics of a specific subset of tasks rather than general ill-suitedness of such models for language generation.
Much research has reported the training data of summarization models are noisy; summaries often do not reflect what is written in the source texts. We propose an effective method of curriculum learning to train summarization models from such noisy data. Curriculum learning is used to train sequence-to-sequence models with noisy data. In translation tasks, previous research quantified noise of the training data using two models trained with noisy and clean corpora. Because such corpora do not exist in summarization fields, we propose a model that can quantify noise from a single noisy corpus. We conduct experiments on three summarization models; one pretrained model and two non-pretrained models, and verify our method improves the performance. Furthermore, we analyze how different curricula affect the performance of pretrained and non-pretrained summarization models. Our result on human evaluation also shows our method improves the performance of summarization models.
Lexical collocations are idiosyncratic combinations of two syntactically bound lexical items (e.g., “heavy rain” or “take a step”). Understanding their degree of compositionality and idiosyncrasy, as well their underlying semantics, is crucial for language learners, lexicographers and downstream NLP applications. In this paper, we perform an exhaustive analysis of current language models for collocation understanding. We first construct a dataset of apparitions of lexical collocations in context, categorized into 17 representative semantic categories. Then, we perform two experiments: (1) unsupervised collocate retrieval using BERT, and (2) supervised collocation classification in context. We find that most models perform well in distinguishing light verb constructions, especially if the collocation’s first argument acts as subject, but often fail to distinguish, first, different syntactic structures within the same semantic category, and second, fine-grained semantic categories which restrict the use of small sets of valid collocates for a given base.
We propose a novel solution for assigning labels to topic models by using multiple weak labelers. The method leverages generative transformers to learn accurate representations of the most important topic terms and candidate labels. This is achieved by fine-tuning pre-trained BART models on a large number of potential labels generated by state of the art non-neural models for topic labeling, enriched with different techniques. The proposed BART-TL model is able to generate valuable and novel labels in a weakly-supervised manner and can be improved by adding other weak labelers or distant supervision on similar tasks.
Textual information extraction is a typical research topic in the NLP community. Several NLP tasks such as named entity recognition and relation extraction between entities have been well-studied in previous work. However, few works pay their attention to the implicit information. For example, a financial news article mentioned “Apple Inc.” may be also related to Samsung, even though Samsung is not explicitly mentioned in this article. This work presents a novel dynamic graph transformer that distills the textual information and the entity relations on the fly. Experimental results confirm the effectiveness of our approach to implicit tag recognition.
Likelihood training and maximization-based decoding result in dull and repetitive generated texts even when using powerful language models (Holtzman et al., 2019). Adding a loss function for regularization was shown to improve text generation output by helping avoid unwanted properties, such as contradiction or repetition (Li at al., 2020). In this work, we propose fine-tuning a language model by using policy gradient reinforcement learning, directly optimizing for better generation. We apply this approach to minimizing repetition in generated text, and show that, when combined with unlikelihood training (Welleck et al., 2020), our method further reduces repetition without impacting the language model quality. We also evaluate other methods for improving generation at training and decoding time, and compare them using various metrics aimed at control for better text generation output.
Platforms that support online commentary, from social networks to news sites, are increasingly leveraging machine learning to assist their moderation efforts. But this process does not typically provide feedback to the author that would help them contribute according to the community guidelines. This is prohibitively time-consuming for human moderators to do, and computational approaches are still nascent. This work focuses on models that can help suggest rephrasings of toxic comments in a more civil manner. Inspired by recent progress in unpaired sequence-to-sequence tasks, a self-supervised learning model is introduced, called CAE-T5. CAE-T5 employs a pre-trained text-to-text transformer, which is fine tuned with a denoising and cyclic auto-encoder loss. Experimenting with the largest toxicity detection dataset to date (Civil Comments) our model generates sentences that are more fluent and better at preserving the initial content compared to earlier text style transfer systems which we compare with using several scoring systems and human evaluation.
The task of generating weather-forecast comments from meteorological simulations has the following requirements: (i) the changes in numerical values for various physical quantities need to be considered, (ii) the weather comments should be dependent on delivery time and area information, and (iii) the comments should provide useful information for users. To meet these requirements, we propose a data-to-text model that incorporates three types of encoders for numerical forecast maps, observation data, and meta-data. We also introduce weather labels representing weather information, such as sunny and rain, for our model to explicitly describe useful information. We conducted automatic and human evaluations. The results indicate that our model performed best against baselines in terms of informativeness. We make our code and data publicly available.
We present SICK-NL (read: signal), a dataset targeting Natural Language Inference in Dutch. SICK-NL is obtained by translating the SICK dataset of (Marelli et al., 2014) from English into Dutch. Having a parallel inference dataset allows us to compare both monolingual and multilingual NLP models for English and Dutch on the two tasks. In the paper, we motivate and detail the translation process, perform a baseline evaluation on both the original SICK dataset and its Dutch incarnation SICK-NL, taking inspiration from Dutch skipgram embeddings and contextualised embedding models. In addition, we encapsulate two phenomena encountered in the translation to formulate stress tests and verify how well the Dutch models capture syntactic restructurings that do not affect semantics. Our main finding is all models perform worse on SICK-NL than on SICK, indicating that the Dutch dataset is more challenging than the English original. Results on the stress tests show that models don’t fully capture word order freedom in Dutch, warranting future systematic studies.
Non-native speakers show difficulties with spoken word processing. Many studies attribute these difficulties to imprecise phonological encoding of words in the lexical memory. We test an alternative hypothesis: that some of these difficulties can arise from the non-native speakers’ phonetic perception. We train a computational model of phonetic learning, which has no access to phonology, on either one or two languages. We first show that the model exhibits predictable behaviors on phone-level and word-level discrimination tasks. We then test the model on a spoken word processing task, showing that phonology may not be necessary to explain some of the word processing effects observed in non-native speakers. We run an additional analysis of the model’s lexical representation space, showing that the two training languages are not fully separated in that space, similarly to the languages of a bilingual human speaker.
The advent of neural-networks in NLP brought with it substantial improvements in supervised relation extraction. However, obtaining a sufficient quantity of training data remains a key challenge. In this work we propose a process for bootstrapping training datasets which can be performed quickly by non-NLP-experts. We take advantage of search engines over syntactic-graphs (Such as Shlain et al. (2020)) which expose a friendly by-example syntax. We use these to obtain positive examples by searching for sentences that are syntactically similar to user input examples. We apply this technique to relations from TACRED and DocRED and show that the resulting models are competitive with models trained on manually annotated data and on data obtained from distant supervision. The models also outperform models trained using NLG data augmentation techniques. Extending the search-based approach with the NLG method further improves the results.
Systems that generate natural language text from abstract meaning representations such as AMR are typically evaluated using automatic surface matching metrics that compare the generated texts to reference texts from which the input meaning representations were constructed. We show that besides well-known issues from which such metrics suffer, an additional problem arises when applying these metrics for AMR-to-text evaluation, since an abstract meaning representation allows for numerous surface realizations. In this work we aim to alleviate these issues by proposing ℳℱ𝛽, a decomposable metric that builds on two pillars. The first is the principle of meaning preservation ℳ: it measures to what extent a given AMR can be reconstructed from the generated sentence using SOTA AMR parsers and applying (fine-grained) AMR evaluation metrics to measure the distance between the original and the reconstructed AMR. The second pillar builds on a principle of (grammatical) form ℱ that measures the linguistic quality of the generated text, which we implement using SOTA language models. In two extensive pilot studies we show that fulfillment of both principles offers benefits for AMR-to-text evaluation, including explainability of scores. Since ℳℱ𝛽 does not necessarily rely on gold AMRs, it may extend to other text generation tasks.
While we live in an increasingly interconnected world, different places still exhibit strikingly different cultures and many events we experience in our every day life pertain only to the specific place we live in. As a result, people often talk about different things in different parts of the world. In this work we study the effect of local context in machine translation and postulate that this causes the domains of the source and target language to greatly mismatch. We first formalize the concept of source-target domain mismatch, propose a metric to quantify it, and provide empirical evidence for its existence. We conclude with an empirical study of how source-target domain mismatch affects training of machine translation systems on low resource languages. While this may severely affect back-translation, the degradation can be alleviated by combining back-translation with self-training and by increasing the amount of target side monolingual data.
There has been much interest in rumor detection using deep learning models in recent years. A well-known limitation of deep learning models is that they tend to learn superficial patterns, which restricts their generalization ability. We find that this is also true for cross-topic rumor detection. In this paper, we propose a method inspired by the “mixture of experts” paradigm. We assume that the prediction of the rumor class label given an instance is dependent on the topic distribution of the instance. After deriving a vector representation for each topic, given an instance, we derive a “topic mixture” vector for the instance based on its topic distribution. This topic mixture is combined with the vector representation of the instance itself to make rumor predictions. Our experiments show that our proposed method can outperform two baseline debiasing methods in a cross-topic setting. In a synthetic setting when we removed topic-specific words, our method also works better than the baselines, showing that our method does not rely on superficial features.
Pre-editing is the process of modifying the source text (ST) so that it can be translated by machine translation (MT) in a better quality. Despite the unpredictability of black-box neural MT (NMT), pre-editing has been deployed in various practical MT use cases. Although many studies have demonstrated the effectiveness of pre-editing methods for particular settings, thus far, a deep understanding of what pre-editing is and how it works for black-box NMT is lacking. To elicit such understanding, we extensively investigated human pre-editing practices. We first implemented a protocol to incrementally record the minimum edits for each ST and collected 6,652 instances of pre-editing across three translation directions, two MT systems, and four text domains. We then analysed the instances from three perspectives: the characteristics of the pre-edited ST, the diversity of pre-editing operations, and the impact of the pre-editing operations on NMT outputs. Our findings include the following: (1) enhancing the explicitness of the meaning of an ST and its syntactic structure is more important for obtaining better translations than making the ST shorter and simpler, and (2) although the impact of pre-editing on NMT is generally unpredictable, there are some tendencies of changes in the NMT outputs depending on the editing operation types.
Embedding entities and relations of a knowledge graph in a low-dimensional space has shown impressive performance in predicting missing links between entities. Although progresses have been achieved, existing methods are heuristically motivated and theoretical understanding of such embeddings is comparatively underdeveloped. This paper extends the random walk model of word embeddings to Knowledge Graph Embeddings (KGEs) to derive a scoring function that evaluates the strength of a relation R between two entities h (head) and t (tail). Moreover, we show that marginal loss minimisation, a popular objective used in much prior work in KGE, follows naturally from the log-likelihood ratio maximisation under the probabilities estimated from the KGEs according to our theoretical relationship. We propose a learning objective motivated by the theoretical analysis to learn KGEs from a given knowledge graph.Using the derived objective, accurate KGEs are learnt from FB15K237 and WN18RR benchmark datasets, providing empirical evidence in support of the theory.
Metric-based learning is a well-known family of methods for few-shot learning, especially in computer vision. Recently, they have been used in many natural language processing applications but not for slot tagging. In this paper, we explore metric-based learning methods in the slot tagging task and propose a novel metric-based learning architecture - Attentive Relational Network. Our proposed method extends relation networks, making them more suitable for natural language processing applications in general, by leveraging pretrained contextual embeddings such as ELMO and BERT and by using attention mechanism. The results on SNIPS data show that our proposed method outperforms other state of the art metric-based learning methods.
Emotion recognition (ER) is an important task in Natural Language Processing (NLP), due to its high impact in real-world applications from health and well-being to author profiling, consumer analysis and security. Current approaches to ER, mainly classify emotions independently without considering that emotions can co-exist. Such approaches overlook potential ambiguities, in which multiple emotions overlap. We propose a new model “SpanEmo” casting multi-label emotion classification as span-prediction, which can aid ER models to learn associations between labels and words in a sentence. Furthermore, we introduce a loss function focused on modelling multiple co-existing emotions in the input sentence. Experiments performed on the SemEval2018 multi-label emotion data over three language sets (i.e., English, Arabic and Spanish) demonstrate our method’s effectiveness. Finally, we present different analyses that illustrate the benefits of our method in terms of improving the model performance and learning meaningful associations between emotion classes and words in the sentence.
Keyphrases associated with research papers provide an effective way to find useful information in the large and growing scholarly digital collections. In this paper, we present KPRank, an unsupervised graph-based algorithm for keyphrase extraction that exploits both positional information and contextual word embeddings into a biased PageRank. Our experimental results on five benchmark datasets show that KPRank that uses contextual word embeddings with additional position signal outperforms previous approaches and strong baselines for this task.
Machine reading comprehension (MRC) has received considerable attention as a benchmark for natural language understanding. However, the conventional task design of MRC lacks explainability beyond the model interpretation, i.e., reading comprehension by a model cannot be explained in human terms. To this end, this position paper provides a theoretical basis for the design of MRC datasets based on psychology as well as psychometrics, and summarizes it in terms of the prerequisites for benchmarking MRC. We conclude that future datasets should (i) evaluate the capability of the model for constructing a coherent and grounded representation to understand context-dependent situations and (ii) ensure substantive validity by shortcut-proof questions and explanation as a part of the task design.
Recent work in multilingual translation advances translation quality surpassing bilingual baselines using deep transformer models with increased capacity. However, the extra latency and memory costs introduced by this approach may make it unacceptable for efficiency-constrained applications. It has recently been shown for bilingual translation that using a deep encoder and shallow decoder (DESD) can reduce inference latency while maintaining translation quality, so we study similar speed-accuracy trade-offs for multilingual translation. We find that for many-to-one translation we can indeed increase decoder speed without sacrificing quality using this approach, but for one-to-many translation, shallow decoders cause a clear quality drop. To ameliorate this drop, we propose a deep encoder with multiple shallow decoders (DEMSD) where each shallow decoder is responsible for a disjoint subset of target languages. Specifically, the DEMSD model with 2-layer decoders is able to obtain a 1.8x speedup on average compared to a standard transformer model with no drop in translation quality.
Sentence Compression is the task of generating a shorter, yet grammatical, version of a given sentence, preserving the essence of the original sentence. This paper proposes a Black-Box Optimizer for Compression (B-BOC): given a black-box compression algorithm and assuming not all sentences need be compressed – find the best candidates for compression in order to maximize both compression rate and quality. Given a required compression ratio, we consider two scenarios: (i) single-sentence compression, and (ii) sentences-sequence compression. In the first scenario our optimizer is trained to predict how well each sentence could be compressed while meeting the specified ratio requirement. In the latter, the desired compression ratio is applied to a sequence of sentences (e.g., a paragraph) as a whole, rather than on each individual sentence. To achieve that we use B-BOC to assign an optimal compression ratio to each sentence, then cast it as a Knapsack problem which we solve using bounded dynamic programming. We evaluate B-BOC on both scenarios on three datasets, demonstrating that our optimizer improves both accuracy and Rouge-F1-score compared to direct application of other compression algorithms.
We present WiC-TSV, a new multi-domain evaluation benchmark for Word Sense Disambiguation. More specifically, we introduce a framework for Target Sense Verification of Words in Context which grounds its uniqueness in the formulation as binary classification task thus being independent of external sense inventories, and the coverage of various domains. This makes the dataset highly flexible for the evaluation of a diverse set of models and systems in and across domains. WiC-TSV provides three different evaluation settings, depending on the input signals provided to the model. We set baseline performance on the dataset using state-of-the-art language models. Experimental results show that even though these models can perform decently on the task, there remains a gap between machine and human performance, especially in out-of-domain settings. WiC-TSV data is available at https://competitions.codalab.org/competitions/23683.
We address the problem of unsupervised abstractive summarization of collections of user generated reviews through self-supervision and control. We propose a self-supervised setup that considers an individual document as a target summary for a set of similar documents. This setting makes training simpler than previous approaches by relying only on standard log-likelihood loss and mainstream models. We address the problem of hallucinations through the use of control codes, to steer the generation towards more coherent and relevant summaries.
Previous research on target-dependent sentiment classification (TSC) has mostly focused on reviews, social media, and other domains where authors tend to express sentiment explicitly. In this paper, we investigate TSC in news articles, a much less researched TSC domain despite the importance of news as an essential information source in individual and societal decision making. We introduce NewsMTSC, a high-quality dataset for TSC on news articles with key differences compared to established TSC datasets, including, for example, different means to express sentiment, longer texts, and a second test-set to measure the influence of multi-target sentences. We also propose a model that uses a BiGRU to interact with multiple embeddings, e.g., from a language model and external knowledge sources. The proposed model improves the performance of the prior state-of-the-art from F1_m=81.7 to 83.1 (real-world sentiment distribution) and from F1_m=81.2 to 82.5 (multi-target sentences).
Many data sets (e.g., reviews, forums, news, etc.) exist parallelly in multiple languages. They all cover the same content, but the linguistic differences make it impossible to use traditional, bag-of-word-based topic models. Models have to be either single-language or suffer from a huge, but extremely sparse vocabulary. Both issues can be addressed by transfer learning. In this paper, we introduce a zero-shot cross-lingual topic model. Our model learns topics on one language (here, English), and predicts them for unseen documents in different languages (here, Italian, French, German, and Portuguese). We evaluate the quality of the topic predictions for the same document in different languages. Our results show that the transferred topics are coherent and stable across languages, which suggests exciting future research directions.
Modern neural approaches to dependency parsing are trained to predict a tree structure by jointly learning a contextual representation for tokens in a sentence, as well as a head–dependent scoring function. Whereas this strategy results in high performance, it is difficult to interpret these representations in relation to the geometry of the underlying tree structure. Our work seeks instead to learn interpretable representations by training a parser to explicitly preserve structural properties of a tree. We do so by casting dependency parsing as a tree embedding problem where we incorporate geometric properties of dependency trees in the form of training losses within a graph-based parser. We provide a thorough evaluation of these geometric losses, showing that a majority of them yield strong tree distance preservation as well as parsing performance on par with a competitive graph-based parser (Qi et al., 2018). Finally, we show where parsing errors lie in terms of tree relationship in order to guide future work.
Annotating training data for sequence tagging of texts is usually very time-consuming. Recent advances in transfer learning for natural language processing in conjunction with active learning open the possibility to significantly reduce the necessary annotation budget. We are the first to thoroughly investigate this powerful combination for the sequence tagging task. We conduct an extensive empirical study of various Bayesian uncertainty estimation methods and Monte Carlo dropout options for deep pre-trained models in the active learning framework and find the best combinations for different types of models. Besides, we also demonstrate that to acquire instances during active learning, a full-size Transformer can be substituted with a distilled version, which yields better computational performance and reduces obstacles for applying deep active learning in practice.
When responding to a disaster, humanitarian experts must rapidly process large amounts of secondary data sources to derive situational awareness and guide decision-making. While these documents contain valuable information, manually processing them is extremely time-consuming when an expedient response is necessary. To improve this process, effective summarization models are a valuable tool for humanitarian response experts as they provide digestible overviews of essential information in secondary data. This paper focuses on extractive summarization for the humanitarian response domain and describes and makes public a new multilingual data collection for this purpose. The collection – called MultiHumES– provides multilingual documents coupled with informative snippets that have been annotated by humanitarian analysts over the past four years. We report the performance results of a recent neural networks-based summarization model together with other baselines. We hope that the released data collection can further grow the research on multilingual extractive summarization in the humanitarian response domain.
Assessing the quality of arguments and of the claims the arguments are composed of has become a key task in computational argumentation. However, even if different claims share the same stance on the same topic, their assessment depends on the prior perception and weighting of the different aspects of the topic being discussed. This renders it difficult to learn topic-independent quality indicators. In this paper, we study claim quality assessment irrespective of discussed aspects by comparing different revisions of the same claim. We compile a large-scale corpus with over 377k claim revision pairs of various types from kialo.com, covering diverse topics from politics, ethics, entertainment, and others. We then propose two tasks: (a) assessing which claim of a revision pair is better, and (b) ranking all versions of a claim by quality. Our first experiments with embedding-based logistic regression and transformer-based neural networks show promising results, suggesting that learned indicators generalize well across topics. In a detailed error analysis, we give insights into what quality dimensions of claims can be assessed reliably. We provide the data and scripts needed to reproduce all results.
Dialogue State Tracking (DST) forms a core component of automated chatbot based systems designed for specific goals like hotel, taxi reservation, tourist information etc. With the increasing need to deploy such systems in new domains, solving the problem of zero/few-shot DST has become necessary. There has been a rising trend for learning to transfer knowledge from resource-rich domains to unknown domains with minimal need for additional data. In this work, we explore the merits of meta-learning algorithms for this transfer and hence, propose a meta-learner D-REPTILE specific to the DST problem. With extensive experimentation, we provide clear evidence of benefits over conventional approaches across different domains, methods, base models and datasets with significant (5-25%) improvement over the baseline in a low-data setting. Our proposed meta-learner is agnostic of the underlying model and hence any existing state-of-the-art DST system can improve its performance on unknown domains using our training strategy.
Pretrained transformer-based models, such as BERT and its variants, have become a common choice to obtain state-of-the-art performances in NLP tasks. In the identification of Adverse Drug Events (ADE) from social media texts, for example, BERT architectures rank first in the leaderboard. However, a systematic comparison between these models has not yet been done. In this paper, we aim at shedding light on the differences between their performance analyzing the results of 12 models, tested on two standard benchmarks. SpanBERT and PubMedBERT emerged as the best models in our evaluation: this result clearly shows that span-based pretraining gives a decisive advantage in the precise recognition of ADEs, and that in-domain language pretraining is particularly useful when the transformer model is trained just on biomedical text from scratch.
Speech disfluencies are prevalent in spontaneous speech. The rising popularity of voice assistants presents a growing need to handle naturally occurring disfluencies. Semantic parsing is a key component for understanding user utterances in voice assistants, yet most semantic parsing research to date focuses on written text. In this paper, we investigate semantic parsing of disfluent speech with the ATIS dataset. We find that a state-of-the-art semantic parser does not seamlessly handle disfluencies. We experiment with adding real and synthetic disfluencies at training time and find that adding synthetic disfluencies not only improves model performance by up to 39% but can also outperform adding real disfluencies in the ATIS dataset.
In this work, we explore joint energy-based model (EBM) training during the finetuning of pretrained text encoders (e.g., Roberta) for natural language understanding (NLU) tasks. Our experiments show that EBM training can help the model reach a better calibration that is competitive to strong baselines, with little or no loss in accuracy. We discuss three variants of energy functions (namely scalar, hidden, and sharp-hidden) that can be defined on top of a text encoder, and compare them in experiments. Due to the discreteness of text data, we adopt noise contrastive estimation (NCE) to train the energy-based model. To make NCE training more effective, we train an auto-regressive noise model with the masked language model (MLM) objective.
In this paper, we challenge the assumption that political ideology is inherently built into text by presenting an investigation into the impact of experiential factors on annotator perceptions of political ideology. We construct an annotated corpus of U.S. political discussion, where in addition to ideology labels for texts, annotators provide information about their political affiliation, exposure to political news, and familiarity with the source domain of discussion, Reddit. We investigate the variability in ideology judgments across annotators, finding evidence that these experiential factors may influence the consistency of how political ideologies are perceived. Finally, we present evidence that understanding how humans perceive and interpret ideology from texts remains a challenging task for state-of-the-art language models, pointing towards potential issues when modeling user experiences that may require more contextual knowledge.
Pretrained language models have been suggested as a possible alternative or complement to structured knowledge bases. However, this emerging LM-as-KB paradigm has so far only been considered in a very limited setting, which only allows handling 21k entities whose name is found in common LM vocabularies. Furthermore, a major benefit of this paradigm, i.e., querying the KB using natural language paraphrases, is underexplored. Here we formulate two basic requirements for treating LMs as KBs: (i) the ability to store a large number facts involving a large number of entities and (ii) the ability to query stored facts. We explore three entity representations that allow LMs to handle millions of entities and present a detailed case study on paraphrased querying of facts stored in LMs, thereby providing a proof-of-concept that language models can indeed serve as knowledge bases.
Fine-tuning a large language model on downstream tasks has become a commonly adopted process in the Natural Language Processing (NLP) (CITATION). However, such a process, when associated with the current transformer-based (CITATION) architectures, shows several limitations when the target task requires to reason with long documents. In this work, we introduce a novel hierarchical propagation layer that spreads information between multiple transformer windows. We adopt a hierarchical approach where the input is divided in multiple blocks independently processed by the scaled dot-attentions and combined between the successive layers. We validate the effectiveness of our approach on three extractive summarization corpora of long scientific papers and news articles. We compare our approach to standard and pre-trained language-model-based summarizers and report state-of-the-art results for long document summarization and comparable results for smaller document summarization.
Given the potential misuse of recent advances in synthetic text generation by language models (LMs), it is important to have the capacity to attribute authorship of synthetic text. While stylometric organic (i.e., human written) authorship attribution has been quite successful, it is unclear whether similar approaches can be used to attribute a synthetic text to its source LM. We address this question with the key insight that synthetic texts carry subtle distinguishing marks inherited from their source LM and that these marks can be leveraged by machine learning (ML) algorithms for attribution. We propose and test several ML-based attribution methods. Our best attributor built using a fine-tuned version of XLNet (XLNet-FT) consistently achieves excellent accuracy scores (91% to near perfect 98%) in terms of attributing the parent pre-trained LM behind a synthetic text. Our experiments show promising results across a range of experiments where the synthetic text may be generated using pre-trained LMs, fine-tuned LMs, or by varying text generation parameters.
(CITATION) argued for using random splits rather than standard splits in NLP experiments. We argue that random splits, like standard splits, lead to overly optimistic performance estimates. We can also split data in biased or adversarial ways, e.g., training on short sentences and evaluating on long ones. Biased sampling has been used in domain adaptation to simulate real-world drift; this is known as the covariate shift assumption. In NLP, however, even worst-case splits, maximizing bias, often under-estimate the error observed on new samples of in-domain data, i.e., the data that models should minimally generalize to at test time. This invalidates the covariate shift assumption. Instead of using multiple random splits, future benchmarks should ideally include multiple, independent test sets instead; if infeasible, we argue that multiple biased splits leads to more realistic performance estimates than multiple random splits.
In this work, we consider the problem of uncertainty estimation for Transformer-based models. We investigate the applicability of uncertainty estimates based on dropout usage at the inference stage (Monte Carlo dropout). The series of experiments on natural language understanding tasks shows that the resulting uncertainty estimates improve the quality of detection of error-prone instances. Special attention is paid to the construction of computationally inexpensive estimates via Monte Carlo dropout and Determinantal Point Processes.
The present article discusses how to improve translation quality when using limited training data to translate towards morphologically rich languages. The starting point is a neural MT system, used to train translation models, using solely publicly available parallel data. An initial analysis of the translation output has shown that quality is sub-optimal, due mainly to an insufficient amount of training data. To improve translation quality, a hybridized solution is proposed, using an ensemble of relatively simple NMT systems trained with different metrics, combined with an open source module, designed for a low-resource MT system. Experimental results of the proposed hybridized method with multiple independent test sets achieve improvements over (i) both the best individual NMT and (ii) the standard ensemble system provided in the Marian-NMT system. Improvements over Marian-NMT are in many cases statistically significant. Finally, a qualitative analysis of translation results indicates a greater robustness for the hybridized method.
We introduce a data augmentation technique based on byte pair encoding and a BERT-like self-attention model to boost performance on spoken language understanding tasks. We compare and evaluate this method with a range of augmentation techniques encompassing generative models such as VAEs and performance-boosting techniques such as synonym replacement and back-translation. We show our method performs strongly on domain and intent classification tasks for a voice assistant and in a user-study focused on utterance naturalness and semantic similarity.
Manual evaluation is essential to judge progress on automatic text summarization. However, we conduct a survey on recent summarization system papers that reveals little agreement on how to perform such evaluation studies. We conduct two evaluation experiments on two aspects of summaries’ linguistic quality (coherence and repetitiveness) to compare Likert-type and ranking annotations and show that best choice of evaluation method can vary from one aspect to another. In our survey, we also find that study parameters such as the overall number of annotators and distribution of annotators to annotation items are often not fully reported and that subsequent statistical analysis ignores grouping factors arising from one annotator judging multiple summaries. Using our evaluation experiments, we show that the total number of annotators can have a strong impact on study power and that current statistical analysis methods can inflate type I error rates up to eight-fold. In addition, we highlight that for the purpose of system comparison the current practice of eliciting multiple judgements per summary leads to less powerful and reliable annotations given a fixed study budget.
Linguistic accommodation is the process in which speakers adjust their accent, diction, vocabulary, and other aspects of language according to the communication style of one another. Previous research has shown how linguistic accommodation correlates with gaps in the power and status of the speakers and the way it promotes approval and discussion efficiency. In this work, we provide a novel perspective on the phenomena, exploring its correlation with the open-mindedness of a speaker, rather than to her social status. We process thousands of unstructured argumentative discussions that took place in Reddit’s Change My View (CMV) subreddit, demonstrating that open-mindedness relates to the assumed role of a speaker in different contexts. On the discussion level, we surprisingly find that discussions that reach agreement present lower levels of accommodation.
We evaluate two common conjectures in error analysis of NLP models: (i) Morphology is predictive of errors; and (ii) the importance of morphology increases with the morphological complexity of a language. We show across four different tasks and up to 57 languages that of these conjectures, somewhat surprisingly, only (i) is true. Using morphological features does improve error prediction across tasks; however, this effect is less pronounced with morphologically complex languages. We speculate this is because morphology is more discriminative in morphologically simple languages. Across all four tasks, case and gender are the morphological features most predictive of error.
The transformer has been shown to outperform recurrent neural network-based sequence-to-sequence models in various word-level NLP tasks. Yet for character-level transduction tasks, e.g. morphological inflection generation and historical text normalization, there are few works that outperform recurrent models using the transformer. In an empirical study, we uncover that, in contrast to recurrent sequence-to-sequence models, the batch size plays a crucial role in the performance of the transformer on character-level tasks, and we show that with a large enough batch size, the transformer does indeed outperform recurrent models. We also introduce a simple technique to handle feature-guided character-level transduction that further improves performance. With these insights, we achieve state-of-the-art performance on morphological inflection and historical text normalization. We also show that the transformer outperforms a strong baseline on two other character-level transduction tasks: grapheme-to-phoneme conversion and transliteration.
Reinforcement Learning (RL) is a powerful framework to address the discrepancy between loss functions used during training and the final evaluation metrics to be used at test time. When applied to neural Machine Translation (MT), it minimises the mismatch between the cross-entropy loss and non-differentiable evaluation metrics like BLEU. However, the suitability of these metrics as reward function at training time is questionable: they tend to be sparse and biased towards the specific words used in the reference texts. We propose to address this problem by making models less reliant on such metrics in two ways: (a) with an entropy-regularised RL method that does not only maximise a reward function but also explore the action space to avoid peaky distributions; (b) with a novel RL method that explores a dynamic unsupervised reward function to balance between exploration and exploitation. We base our proposals on the Soft Actor-Critic (SAC) framework, adapting the off-policy maximum entropy model for language generation applications such as MT. We demonstrate that SAC with BLEU reward tends to overfit less to the training data and performs better on out-of-domain data. We also show that our dynamic unsupervised reward can lead to better translation of ambiguous words.
Computational modelling of political discourse tasks has become an increasingly important area of research in the field of natural language processing. Populist rhetoric has risen across the political sphere in recent years; however, due to its complex nature, computational approaches to it have been scarce. In this paper, we present the new Us vs. Them dataset, consisting of 6861 Reddit comments annotated for populist attitudes and the first large-scale computational models of this phenomenon. We investigate the relationship between populist mindsets and social groups, as well as a range of emotions typically associated with these. We set a baseline for two tasks associated with populist attitudes and present a set of multi-task learning models that leverage and demonstrate the importance of emotion and group identification as auxiliary tasks.
We present a novel dataset and model for a multilingual setting to approach the task of Joint Entity and Relation Extraction. The SMiLER dataset consists of 1.1 M annotated sentences, representing 36 relations, and 14 languages. To the best of our knowledge, this is currently both the largest and the most comprehensive dataset of this type. We introduce HERBERTa, a pipeline that combines two independent BERT models: one for sequence classification, and the other for entity tagging. The model achieves micro F1 81.49 for English on this dataset, which is close to the current SOTA on CoNLL, SpERT.
Imposing the style of one image onto another is called style transfer. For example, the style of a Van Gogh painting might be imposed on a photograph to yield an interesting hybrid. This paper applies the adaptive normalization used for image style transfer to language semantics, i.e., the style is the way the words are said (tone of voice and facial expressions) and these are style-transferred onto the text. The goal is to learn richer representations for multi-modal utterances using style-transferred multi-modal features. The proposed Style-Transfer Transformer (STT) grafts a stepped styled adaptive layer-normalization onto a transformer network, the output from which is used in sentiment analysis and emotion recognition problems. In addition to achieving performance on par with the state-of-the art (but using less than a third of the model parameters), we examine the relative contributions of each mode when used in the downstream applications.
The great majority of languages in the world are considered under-resourced for successful application of deep learning methods. In this work, we propose a meta-learning approach to document classification in low-resource languages and demonstrate its effectiveness in two different settings: few-shot, cross-lingual adaptation to previously unseen languages; and multilingual joint-training when limited target-language data is available during trai-ing. We conduct a systematic comparison of several meta-learning methods, investigate multiple settings in terms of data availability, and show that meta-learning thrives in settings with a heterogeneous task distribution. We propose a simple, yet effective adjustment to existing meta-learning methods which allows for better and more stable learning, and set a new state-of-the-art on a number of languages while performing on-par on others, using only a small amount of labeled data.
Biomedical question-answering (QA) has gained increased attention for its capability to provide users with high-quality information from a vast scientific literature. Although an increasing number of biomedical QA datasets has been recently made available, those resources are still rather limited and expensive to produce; thus, transfer learning via pre-trained language models (LMs) has been shown as a promising approach to leverage existing general-purpose knowledge. However, fine-tuning these large models can be costly and time consuming and often yields limited benefits when adapting to specific themes of specialised domains, such as the COVID-19 literature. Therefore, to bootstrap further their domain adaptation, we propose a simple yet unexplored approach, which we call biomedical entity-aware masking (BEM) strategy, encouraging masked language models to learn entity-centric knowledge based on the pivotal entities characterizing the domain at hand, and employ those entities to drive the LM fine-tuning. The resulting strategy is a downstream process applicable to a wide variety of masked LMs, not requiring additional memory or components in the neural architectures. Experimental results show performance on par with the state-of-the-art models on several biomedical QA datasets.
Target-oriented opinion words extraction (TOWE) is a subtask of aspect-based sentiment analysis (ABSA). It aims to extract the corresponding opinion words for a given opinion target in a review sentence. Intuitively, the relation between an opinion target and an opinion word mostly relies on syntactics. In this study, we design a directed syntactic dependency graph based on a dependency tree to establish a path from the target to candidate opinions. Subsequently, we propose a novel attention-based relational graph convolutional neural network (ARGCN) to exploit syntactic information over dependency graphs. Moreover, to explicitly extract the corresponding opinion words toward the given opinion target, we effectively encode target information in our model with the target-aware representation. Empirical results demonstrate that our model significantly outperforms all of the existing models on four benchmark datasets. Extensive analysis also demonstrates the effectiveness of each component of our models. Our code is available at https://github.com/wcwowwwww/towe-eacl.
Detecting arguments in online interactions is useful to understand how conflicts arise and get resolved. Users often use figurative language, such as sarcasm, either as persuasive devices or to attack the opponent by an ad hominem argument. To further our understanding of the role of sarcasm in shaping the disagreement space, we present a thorough experimental setup using a corpus annotated with both argumentative moves (agree/disagree) and sarcasm. We exploit joint modeling in terms of (a) applying discrete features that are useful in detecting sarcasm to the task of argumentative relation classification (agree/disagree/none), and (b) multitask learning for argumentative relation classification and sarcasm detection using deep learning architectures (e.g., dual Long Short-Term Memory (LSTM) with hierarchical attention and Transformer-based architectures). We demonstrate that modeling sarcasm improves the argumentative relation classification task (agree/disagree/none) in all setups.
Relation schemas are often pre-defined for each relation dataset. Relation types can be related from different datasets and have overlapping semantics. We hypothesize we can combine these datasets according to the semantic relatedness between the relation types to overcome the problem of lack of training data. It is often easy to discover the connection between relation types based on relation names or annotation guides, but hard to measure the exact similarity and take advantage of the connection between the relation types from different datasets. We propose to use prototypical examples to represent each relation type and use these examples to augment related types from a different dataset. We obtain further improvement (ACE05) with this type augmentation over a strong baseline which uses multi-task learning between datasets to obtain better feature representation for relations. We make our implementation publicly available: https://github.com/fufrank5/relatedness
Disagreements are pervasive in human communication. In this paper we investigate what makes disagreement constructive. To this end, we construct WikiDisputes, a corpus of 7425 Wikipedia Talk page conversations that contain content disputes, and define the task of predicting whether disagreements will be escalated to mediation by a moderator. We evaluate feature-based models with linguistic markers from previous work, and demonstrate that their performance is improved by using features that capture changes in linguistic markers throughout the conversations, as opposed to averaged values. We develop a variety of neural models and show that taking into account the structure of the conversation improves predictive accuracy, exceeding that of feature-based models. We assess our best neural model in terms of both predictive accuracy and uncertainty by evaluating its behaviour when it is only exposed to the beginning of the conversation, finding that model accuracy improves and uncertainty reduces as models are exposed to more information.
To track different levels of formality in written discourse, we introduce a novel type of lexicon for the German language, with entries ordered by their degree of (in)formality. We start with a set of words extracted from traditional lexicographic resources, extend it by sentence-based similarity computations, and let crowdworkers assess the enlarged set of lexical items on a continuous informal-formal scale as a gold standard for evaluation. We submit this lexicon to an intrinsic evaluation related to the best regression models and their effect on predicting formality scores and complement our investigation by an extrinsic evaluation of formality on a German-language email corpus.
This paper proposes a new task regarding event reason extraction from document-level texts. Unlike the previous causality detection task, we do not assign target events in the text, but only provide structural event descriptions, and such settings accord more with practice scenarios. Moreover, we annotate a large dataset FinReason for evaluation, which provides Reasons annotation for Financial events in company announcements. This task is challenging because the cases of multiple-events, multiple-reasons, and implicit-reasons are included. In total, FinReason contains 8,794 documents, 12,861 financial events and 11,006 reason spans. We also provide the performance of existing canonical methods in event extraction and machine reading comprehension on this task. The results show a 7 percentage point F1 score gap between the best model and human performance, and existing methods are far from resolving this problem.
In this paper, we propose to study language modelling as a multi-task problem, bringing together three strands of research: multi-task learning, linguistics, and interpretability. Based on hypotheses derived from linguistic theory, we investigate whether language models adhere to learning principles of multi-task learning during training. To showcase the idea, we analyse the generalisation behaviour of language models as they learn the linguistic concept of Negative Polarity Items (NPIs). Our experiments demonstrate that a multi-task setting naturally emerges within the objective of the more general task of language modelling. We argue that this insight is valuable for multi-task learning, linguistics and interpretability research and can lead to exciting new findings in all three domains.
Conversational systems enable numerous valuable applications, and question-answering is an important component underlying many of these. However, conversational question-answering remains challenging due to the lack of realistic, domain-specific training data. Inspired by this bottleneck, we focus on conversational question generation as a means to generate synthetic conversations for training and evaluation purposes. We present a number of novel strategies to improve conversational flow and accommodate varying question types and overall fluidity. Specifically, we design ChainCQG as a two-stage architecture that learns question-answer representations across multiple dialogue turns using a flow propagation training strategy. ChainCQG significantly outperforms both answer-aware and answer-unaware SOTA baselines (e.g., up to 48% BLEU-1 improvement). Additionally, our model is able to generate different types of questions, with improved fluidity and coreference alignment.
When training a model on referential dialogue guessing games, the best model is usually chosen based on its task success. We show that in the popular end-to-end approach, this choice prevents the model from learning to generate linguistically richer dialogues, since the acquisition of language proficiency takes longer than learning the guessing task. By comparing models playing different games (GuessWhat, GuessWhich, and Mutual Friends), we show that this discrepancy is model- and task-agnostic. We investigate whether and when better language quality could lead to higher task success. We show that in GuessWhat, models could increase their accuracy if they learn to ground, encode, and decode also words that do not occur frequently in the training set.
Abusive language in online discourse negatively affects a large number of social media users. Many computational methods have been proposed to address this issue of online abuse. The existing work, however, tends to focus on detecting the more explicit forms of abuse leaving the subtler forms of abuse largely untouched. Our work addresses this gap by making three core contributions. First, inspired by the theory of impoliteness, we propose a novel task of detecting a subtler form of abuse, namely unpalatable questions. Second, we publish a context-aware dataset for the task using data from a diverse set of Reddit communities. Third, we implement a wide array of learning models and also investigate the benefits of incorporating conversational context into computational models. Our results show that modeling subtle abuse is feasible but difficult due to the language involved being highly nuanced and context-sensitive. We hope that future research in the field will address such subtle forms of abuse since their harm currently passes unnoticed through existing detection systems.
We study a search-based paraphrase generation scheme where candidate paraphrases are generated by iterated transformations from the original sentence and evaluated in terms of syntax quality, semantic distance, and lexical distance. The semantic distance is derived from BERT, and the lexical quality is based on GPT2 perplexity. To solve this multi-objective search problem, we propose two algorithms: Monte-Carlo Tree Search For Paraphrase Generation (MCPG) and Pareto Tree Search (PTS). We provide an extensive set of experiments on 5 datasets with a rigorous reproduction and validation for several state-of-the-art paraphrase generation algorithms. These experiments show that, although being non explicitly supervised, our algorithms perform well against these baselines.
Word alignment over parallel corpora has a wide variety of applications, including learning translation lexicons, cross-lingual transfer of language processing tools, and automatic evaluation or analysis of translation outputs. The great majority of past work on word alignment has worked by performing unsupervised learning on parallel text. Recently, however, other work has demonstrated that pre-trained contextualized word embeddings derived from multilingually trained language models (LMs) prove an attractive alternative, achieving competitive results on the word alignment task even in the absence of explicit training on parallel data. In this paper, we examine methods to marry the two approaches: leveraging pre-trained LMs but fine-tuning them on parallel text with objectives designed to improve alignment quality, and proposing methods to effectively extract alignments from these fine-tuned models. We perform experiments on five language pairs and demonstrate that our model can consistently outperform previous state-of-the-art models of all varieties. In addition, we demonstrate that we are able to train multilingual word aligners that can obtain robust performance on different language pairs.
Many types of distributional word embeddings (weakly) encode linguistic regularities as directions (the difference between jump and jumped will be in a similar direction to that of walk and walked, and so on). Several attempts have been made to explain this fact. We respond to Allen and Hospedales’ recent (ICML, 2019) theoretical explanation, which claims that word2vec and GloVe will encode linguistic regularities whenever a specific relation of paraphrase holds between the four words involved in the regularity. We demonstrate that the explanation does not go through: the paraphrase relations needed under this explanation do not hold empirically
Guessing games are a prototypical instance of the “learning by interacting” paradigm. This work investigates how well an artificial agent can benefit from playing guessing games when later asked to perform on novel NLP downstream tasks such as Visual Question Answering (VQA). We propose two ways to exploit playing guessing games: 1) a supervised learning scenario in which the agent learns to mimic successful guessing games and 2) a novel way for an agent to play by itself, called Self-play via Iterated Experience Learning (SPIEL). We evaluate the ability of both procedures to generalise: an in-domain evaluation shows an increased accuracy (+7.79) compared with competitors on the evaluation suite CompGuessWhat?!; a transfer evaluation shows improved performance for VQA on the TDIUC dataset in terms of harmonic average accuracy (+5.31) thanks to more fine-grained object representations learned via SPIEL.
Ideological attitudes and stance are often expressed through subtle meanings of words and phrases. Understanding these connotations is critical to recognizing the cultural and emotional perspectives of the speaker. In this paper, we use distant labeling to create a new lexical resource representing connotation aspects for nouns and adjectives. Our analysis shows that it aligns well with human judgments. Additionally, we present a method for creating lexical representations that capture connotations within the embedding space and show that using the embeddings provides a statistically significant improvement on the task of stance detection when data is limited.
Designing profitable trading strategies is complex as stock movements are highly stochastic; the market is influenced by large volumes of noisy data across diverse information sources like news and social media. Prior work mostly treats stock movement prediction as a regression or classification task and is not directly optimized towards profit-making. Further, they do not model the fine-grain temporal irregularities in the release of vast volumes of text that the market responds to quickly. Building on these limitations, we propose a novel hierarchical, learning to rank approach that uses textual data to make time-aware predictions for ranking stocks based on expected profit. Our approach outperforms state-of-the-art methods by over 8% in terms of cumulative profit and risk-adjusted returns in trading simulations on two benchmarks: English tweets and Chinese financial news spanning two major stock indexes and four global markets. Through ablative and qualitative analyses, we build the case for our method as a tool for daily stock trading.
Most well-established data collection methods currently adopted in NLP depend on the as- sumption of speaker literacy. Consequently, the collected corpora largely fail to represent swathes of the global population, which tend to be some of the most vulnerable and marginalised people in society, and often live in rural developing areas. Such underrepresented groups are thus not only ignored when making modeling and system design decisions, but also prevented from benefiting from development outcomes achieved through data-driven NLP. This paper aims to address the under-representation of illiterate communities in NLP corpora: we identify potential biases and ethical issues that might arise when collecting data from rural communities with high illiteracy rates in Low-Income Countries, and propose a set of practical mitigation strategies to help future work.
We develop Process Execution Graphs (PEG), a document-level representation of real-world wet lab biochemistry protocols, addressing challenges such as cross-sentence relations, long-range coreference, grounding, and implicit arguments. We manually annotate PEGs in a corpus of complex lab protocols with a novel interactive textual simulator that keeps track of entity traits and semantic constraints during annotation. We use this data to develop graph-prediction models, finding them to be good at entity identification and local relation extraction, while our corpus facilitates further exploration of challenging long-range relations.
Recent studies in the field of Machine Translation (MT) and Natural Language Processing (NLP) have shown that existing models amplify biases observed in the training data. The amplification of biases in language technology has mainly been examined with respect to specific phenomena, such as gender bias. In this work, we go beyond the study of gender in MT and investigate how bias amplification might affect language in a broader sense. We hypothesize that the ‘algorithmic bias’, i.e. an exacerbation of frequently observed patterns in combination with a loss of less frequent ones, not only exacerbates societal biases present in current datasets but could also lead to an artificially impoverished language: ‘machine translationese’. We assess the linguistic richness (on a lexical and morphological level) of translations created by different data-driven MT paradigms – phrase-based statistical (PB-SMT) and neural MT (NMT). Our experiments show that there is a loss of lexical and syntactic richness in the translations produced by all investigated MT paradigms for two language pairs (EN-FR and EN-ES).
Multilingual pretrained language models have demonstrated remarkable zero-shot cross-lingual transfer capabilities. Such transfer emerges by fine-tuning on a task of interest in one language and evaluating on a distinct language, not seen during the fine-tuning. Despite promising results, we still lack a proper understanding of the source of this transfer. Using a novel layer ablation technique and analyses of the model’s internal representations, we show that multilingual BERT, a popular multilingual language model, can be viewed as the stacking of two sub-networks: a multilingual encoder followed by a task-specific language-agnostic predictor. While the encoder is crucial for cross-lingual transfer and remains mostly unchanged during fine-tuning, the task predictor has little importance on the transfer and can be reinitialized during fine-tuning. We present extensive experiments with three distinct tasks, seventeen typologically diverse languages and multiple domains to support our hypothesis.
This paper proposes two intuitive metrics, skew and stereotype, that quantify and analyse the gender bias present in contextual language models when tackling the WinoBias pronoun resolution task. We find evidence that gender stereotype correlates approximately negatively with gender skew in out-of-the-box models, suggesting that there is a trade-off between these two forms of bias. We investigate two methods to mitigate bias. The first approach is an online method which is effective at removing skew at the expense of stereotype. The second, inspired by previous work on ELMo, involves the fine-tuning of BERT using an augmented gender-balanced dataset. We show that this reduces both skew and stereotype relative to its unaugmented fine-tuned counterpart. However, we find that existing gender bias benchmarks do not fully probe professional bias as pronoun resolution may be obfuscated by cross-correlations from other manifestations of gender prejudice.
The adaptation of pretrained language models to solve supervised tasks has become a baseline in NLP, and many recent works have focused on studying how linguistic information is encoded in the pretrained sentence representations. Among other information, it has been shown that entire syntax trees are implicitly embedded in the geometry of such models. As these models are often fine-tuned, it becomes increasingly important to understand how the encoded knowledge evolves along the fine-tuning. In this paper, we analyze the evolution of the embedded syntax trees along the fine-tuning process of BERT for six different tasks, covering all levels of the linguistic structure. Experimental results show that the encoded syntactic information is forgotten (PoS tagging), reinforced (dependency and constituency parsing) or preserved (semantics-related tasks) in different ways along the fine-tuning process depending on the task.
Integrating external knowledge into commonsense reasoning tasks has shown progress in resolving some, but not all, knowledge gaps in these tasks. For knowledge integration to yield peak performance, it is critical to select a knowledge graph (KG) that is well-aligned with the given task’s objective. We present an approach to assess how well a candidate KG can correctly identify and accurately fill in gaps of reasoning for a task, which we call KG-to-task match. We show this KG-to-task match in 3 phases: knowledge-task identification, knowledge-task alignment, and knowledge-task integration. We also analyze our transformer-based KG-to-task models via commonsense probes to measure how much knowledge is captured in these models before and after KG integration. Empirically, we investigate KG matches for the SocialIQA (SIQA) (Sap et al., 2019b), Physical IQA (PIQA) (Bisk et al., 2020), and MCScript2.0 (Ostermann et al., 2019) datasets with 3 diverse KGs: ATOMIC (Sap et al., 2019a), ConceptNet (Speer et al., 2017), and an automatically constructed instructional KG based on WikiHow (Koupaee and Wang, 2018). With our methods we are able to demonstrate that ATOMIC, an event-inference focused KG, is the best match for SIQA and MCScript2.0, and that the taxonomic ConceptNet and WikiHow-based KGs are the best match for PIQA across all 3 analysis phases. We verify our methods and findings with human evaluation.
It is shown that the optimal next step of an arc-eager parser relative to a non-projective dependency structure can be calculated in cubic time, solving an open problem in parsing theory. Applications are in training of parsers by means of a ‘dynamic oracle’.
Contextual word-representations became a standard in modern natural language processing systems. These models use subword tokenization to handle large vocabularies and unknown words. Word-level usage of such systems requires a way of pooling multiple subwords that correspond to a single word. In this paper we investigate how the choice of subword pooling affects the downstream performance on three tasks: morphological probing, POS tagging and NER, in 9 typologically diverse languages. We compare these in two massively multilingual models, mBERT and XLM-RoBERTa. For morphological tasks, the widely used ‘choose the first subword’ is the worst strategy and the best results are obtained by using attention over the subwords. For POS tagging both of these strategies perform poorly and the best choice is to use a small LSTM over the subwords. The same strategy works best for NER and we show that mBERT is better than XLM-RoBERTa in all 9 languages. We publicly release all code, data and the full result tables at https://github.com/juditacs/subword-choice .
We explore the task of quotability identification, in which, given a document, we aim to identify which of its passages are the most quotable, i.e. the most likely to be directly quoted by later derived documents. We approach quotability identification as a passage ranking problem and evaluate how well both feature-based and BERT-based (Devlin et al., 2019) models rank the passages in a given document by their predicted quotability. We explore this problem through evaluations on five datasets that span multiple languages (English, Latin) and genres of literature (e.g. poetry, plays, novels) and whose corresponding derived documents are of multiple types (news, journal articles). Our experiments confirm the relatively strong performance of BERT-based models on this task, with the best model, a RoBERTA sequential sentence tagger, achieving an average rho of 0.35 and NDCG@1, 5, 50 of 0.26, 0.31 and 0.40, respectively, across all five datasets.
Recent advances in language and vision push forward the research of captioning a single image to describing visual differences between image pairs. Suppose there are two images, I_1 and I_2, and the task is to generate a description W_1,2 comparing them, existing methods directly model I_1, I_2 -> W_1,2 mapping without the semantic understanding of individuals. In this paper, we introduce a Learning-to-Compare (L2C) model, which learns to understand the semantic structures of these two images and compare them while learning to describe each one. We demonstrate that L2C benefits from a comparison between explicit semantic representations and single-image captions, and generalizes better on the new testing image pairs. It outperforms the baseline on both automatic evaluation and human evaluation for the Birds-to-Words dataset.
Voice assistants, e.g., Alexa or Google Assistant, have dramatically improved in recent years. Supporting voice-based search, exploration, and refinement are fundamental tasks for voice assistants, and remain an open challenge. For example, when using voice to search an online shopping site, a user often needs to refine their search by some aspect or facet. This common user intent is usually available through a “filter-by” interface on online shopping websites, but is challenging to support naturally via voice, as the intent of refinements must be interpreted in the context of the original search, the initial results, and the available product catalogue facets. To our knowledge, no benchmark dataset exists for training or validating such contextual search understanding models. To bridge this gap, we introduce the first large-scale dataset of voice-based search refinements, VoiSeR, consisting of about 10,000 search refinement utterances, collected using a novel crowdsourcing task. These utterances are intended to refine a previous search, with respect to a search facet or attribute (e.g., brand, color, review rating, etc.), and are manually annotated with the specific intent. This paper reports qualitative and empirical insights into the most common and challenging types of refinements that a voice-based conversational search system must support. As we show, VoiSeR can support research in conversational query understanding, contextual user intent prediction, and other conversational search topics to facilitate the development of conversational search systems.
We propose a method for online news stream clustering that is a variant of the non-parametric streaming K-means algorithm. Our model uses a combination of sparse and dense document representations, aggregates document-cluster similarity along these multiple representations and makes the clustering decision using a neural classifier. The weighted document-cluster similarity model is learned using a novel adaptation of the triplet loss into a linear classification objective. We show that the use of a suitable fine-tuning objective and external knowledge in pre-trained transformer models yields significant improvements in the effectiveness of contextual embeddings for clustering. Our model achieves a new state-of-the-art on a standard stream clustering dataset of English documents.
In this paper, we propose the Brand-Topic Model (BTM) which aims to detect brand-associated polarity-bearing topics from product reviews. Different from existing models for sentiment-topic extraction which assume topics are grouped under discrete sentiment categories such as ‘positive’, ‘negative’ and ‘neural’, BTM is able to automatically infer real-valued brand-associated sentiment scores and generate fine-grained sentiment-topics in which we can observe continuous changes of words under a certain topic (e.g., ‘shaver’ or ‘cream’) while its associated sentiment gradually varies from negative to positive. BTM is built on the Poisson factorisation model with the incorporation of adversarial learning. It has been evaluated on a dataset constructed from Amazon reviews. Experimental results show that BTM outperforms a number of competitive baselines in brand ranking, achieving a better balance of topic coherence and unique-ness, and extracting better-separated polarity-bearing topics.
Lexical normalization, the translation of non-canonical data to standard language, has shown to improve the performance of many natural language processing tasks on social media. Yet, using multiple languages in one utterance, also called code-switching (CS), is frequently overlooked by these normalization systems, despite its common use in social media. In this paper, we propose three normalization models specifically designed to handle code-switched data which we evaluate for two language pairs: Indonesian-English and Turkish-German. For the latter, we introduce novel normalization layers and their corresponding language ID and POS tags for the dataset, and evaluate the downstream effect of normalization on POS tagging. Results show that our CS-tailored normalization models significantly outperform monolingual ones, and lead to 5.4% relative performance increase for POS tagging as compared to unnormalized input.
Growing concern with online misinformation has encouraged NLP research on fact verification. Since writers often base their assertions on structured data, we focus here on verifying textual statements given evidence in tables. Starting from the Table Parsing (TAPAS) model developed for question answering (Herzig et al., 2020), we find that modeling table structure improves a language model pre-trained on unstructured text. Pre-training language models on English Wikipedia table data further improves performance. Pre-training on a question answering task with column-level cell rank information achieves the best performance. With improved pre-training and cell embeddings, this approach outperforms the state-of-the-art Numerically-aware Graph Neural Network table fact verification model (GNN-TabFact), increasing statement classification accuracy from 72.2% to 73.9% even without modeling numerical information. Incorporating numerical information with cell rankings and pre-training on a question-answering task increases accuracy to 76%. We further analyze accuracy on statements implicating single rows or multiple rows and columns of tables, on different numerical reasoning subtasks, and on generalizing to detecting errors in statements derived from the ToTTo table-to-text generation dataset.
As transparency becomes key for robotics and AI, it will be necessary to evaluate the methods through which transparency is provided, including automatically generated natural language (NL) explanations. Here, we explore parallels between the generation of such explanations and the much-studied field of evaluation of Natural Language Generation (NLG). Specifically, we investigate which of the NLG evaluation measures map well to explanations. We present the ExBAN corpus: a crowd-sourced corpus of NL explanations for Bayesian Networks. We run correlations comparing human subjective ratings with NLG automatic measures. We find that embedding-based automatic NLG evaluation methods, such as BERTScore and BLEURT, have a higher correlation with human ratings, compared to word-overlap metrics, such as BLEU and ROUGE. This work has implications for Explainable AI and transparent robotic and autonomous systems.
Written language contains stylistic cues that can be exploited to automatically infer a variety of potentially sensitive author information. Adversarial stylometry intends to attack such models by rewriting an author’s text. Our research proposes several components to facilitate deployment of these adversarial attacks in the wild, where neither data nor target models are accessible. We introduce a transformer-based extension of a lexical replacement attack, and show it achieves high transferability when trained on a weakly labeled corpus—decreasing target model performance below chance. While not completely inconspicuous, our more successful attacks also prove notably less detectable by humans. Our framework therefore provides a promising direction for future privacy-preserving adversarial attacks.
Much work in cross-lingual transfer learning explored how to select better transfer languages for multilingual tasks, primarily focusing on typological and genealogical similarities between languages. We hypothesize that these measures of linguistic proximity are not enough when working with pragmatically-motivated tasks, such as sentiment analysis. As an alternative, we introduce three linguistic features that capture cross-cultural similarities that manifest in linguistic patterns and quantify distinct aspects of language pragmatics: language context-level, figurative language, and the lexification of emotion concepts. Our analyses show that the proposed pragmatic features do capture cross-cultural similarities and align well with existing work in sociolinguistics and linguistic anthropology. We further corroborate the effectiveness of pragmatically-driven transfer in the downstream task of choosing transfer languages for cross-lingual sentiment analysis.
Recent psychological studies indicate that individuals exhibiting suicidal ideation increasingly turn to social media rather than mental health practitioners. Contextualizing the build-up of such ideation is critical for the identification of users at risk. In this work, we focus on identifying suicidal intent in tweets by augmenting linguistic models with emotional phases modeled from users’ historical context. We propose PHASE, a time-and phase-aware framework that adaptively learns features from a user’s historical emotional spectrum on Twitter for preliminary screening of suicidal risk. Building on clinical studies, PHASE learns phase-like progressions in users’ historical Plutchik-wheel-based emotions to contextualize suicidal intent. While outperforming state-of-the-art methods, we show the utility of temporal and phase-based emotional contextual cues for suicide ideation detection. We further discuss practical and ethical considerations.
Frame identification is one of the key challenges for frame-semantic parsing. The goal of this task is to determine which frame best captures the meaning of a target word or phrase in a sentence. We present a new model for frame identification that uses a pre-trained transformer model to generate representations for frames and lexical units (senses) using their formal definitions in FrameNet. Our frame identification model assesses the suitability of a frame for a target word in a sentence based on the semantic coherence of their meanings. We evaluate our model on three data sets and show that it consistently achieves better performance than previous systems.
Privacy is an important concern when building statistical models on data containing personal information. Differential privacy offers a strong definition of privacy and can be used to solve several privacy concerns. Multiple solutions have been proposed for the differentially-private transformation of datasets containing sensitive information. However, such transformation algorithms offer poor utility in Natural Language Processing (NLP) tasks due to noise added in the process. This paper addresses this issue by providing a utility-preserving differentially private text transformation algorithm using auto-encoders. Our algorithm transforms text to offer robustness against attacks and produces transformations with high semantic quality that perform well on downstream NLP tasks. We prove our algorithm’s theoretical privacy guarantee and assess its privacy leakage under Membership Inference Attacks (MIA) on models trained with transformed data. Our results show that the proposed model performs better against MIA attacks while offering lower to no degradation in the utility of the underlying transformation process compared to existing baselines.
Effective representation of biomedical names for downstream NLP tasks requires the encoding of both lexical as well as domain-specific semantic information. Ideally, the synonymy and semantic relatedness of names should be consistently reflected by their closeness in an embedding space. To achieve such robustness, prior research has considered multi-task objectives when training neural encoders. In this paper, we take a next step towards truly robust representations, which capture more domain-specific semantics while remaining universally applicable across different biomedical corpora and domains. To this end, we use conceptual grounding constraints which more effectively align encoded names to pretrained embeddings of their concept identifiers. These constraints are effective even when using a Deep Averaging Network, a simple feedforward encoding architecture that allows for scaling to large corpora while remaining sufficiently expressive. We empirically validate our approach using multiple tasks and benchmarks, which assess both literal synonymy as well as more general semantic relatedness.
Probabilistic topic models in low data resource scenarios are faced with less reliable estimates due to sparsity of discrete word co-occurrence counts, and do not have the luxury of retraining word or topic embeddings using neural methods. In this challenging resource constrained setting, we explore mixture models which interpolate between the discrete and continuous topic-word distributions that utilise pre-trained embeddings to improve topic coherence. We introduce an automatic trade-off between the discrete and continuous representations via an adaptive mixture coefficient, which places greater weight on the discrete representation when the corpus statistics are more reliable. The adaptive mixture coefficient takes into account global corpus statistics, and the uncertainty in each topic’s continuous distributions. Our approach outperforms the fully discrete, fully continuous, and static mixture model on topic coherence in low resource settings. We additionally demonstrate the generalisability of our method by extending it to handle multilingual document collections.
We evaluate neural model robustness to adversarial attacks using different types of linguistic unit perturbations – character and word, and propose a new method for strategic sentence-level perturbations. We experiment with different amounts of perturbations to examine model confidence and misclassification rate, and contrast model performance with different embeddings BERT and ELMo on two benchmark datasets SQuAD and TriviaQA. We demonstrate how to improve model performance during an adversarial attack by using ensembles. Finally, we analyze factors that effect model behavior under adversarial attack, and develop a new model to predict errors during attacks. Our novel findings reveal that (a) unlike BERT, models that use ELMo embeddings are more susceptible to adversarial attacks, (b) unlike word and paraphrase, character perturbations affect the model the most but are most easily compensated for by adversarial training, (c) word perturbations lead to more high-confidence misclassifications compared to sentence- and character-level perturbations, (d) the type of question and model answer length (the longer the answer the more likely it is to be incorrect) is the most predictive of model errors in adversarial setting, and (e) conclusions about model behavior are dataset-specific.
Automatic unreliable news detection is a research problem with great potential impact. Recently, several papers have shown promising results on large-scale news datasets with models that only use the article itself without resorting to any fact-checking mechanism or retrieving any supporting evidence. In this work, we take a closer look at these datasets. While they all provide valuable resources for future research, we observe a number of problems that may lead to results that do not generalize in more realistic settings. Specifically, we show that selection bias during data collection leads to undesired artifacts in the datasets. In addition, while most systems train and predict at the level of individual articles, overlapping article sources in the training and evaluation data can provide a strong confounding factor that models can exploit. In the presence of this confounding factor, the models can achieve good performance by directly memorizing the site-label mapping instead of modeling the real task of unreliable news detection. We observed a significant drop (>10%) in accuracy for all models tested in a clean split with no train/test source overlap. Using the observations and experimental results, we provide practical suggestions on how to create more reliable datasets for the unreliable news detection task. We suggest future dataset creation include a simple model as a difficulty/bias probe and future model development use a clean non-overlapping site and date split.
Significant memory and computational requirements of large deep neural networks restricts their application on edge devices. Knowledge distillation (KD) is a prominent model compression technique for deep neural networks in which the knowledge of a trained large teacher model is transferred to a smaller student model. The success of knowledge distillation is mainly attributed to its training objective function, which exploits the soft-target information (also known as “dark knowledge”) besides the given regular hard labels in a training set. However, it is shown in the literature that the larger the gap between the teacher and the student networks, the more difficult is their training using knowledge distillation. To address this shortcoming, we propose an improved knowledge distillation method (called Annealing-KD) by feeding the rich information provided by teacher’s soft-targets incrementally and more efficiently. Our Annealing-KD technique is based on a gradual transition over annealed soft-targets generated by the teacher at different temperatures in an iterative process; and therefore, the student is trained to follow the annealed teacher output in a step-by-step manner. This paper includes theoretical and empirical evidence as well as practical experiments to support the effectiveness of our Annealing-KD method. We did a comprehensive set of experiments on different tasks such as image classification (CIFAR-10 and 100) and NLP language inference with BERT-based models on the GLUE benchmark and consistently got superior results.
Unsupervised approaches to extractive summarization usually rely on a notion of sentence importance defined by the semantic similarity between a sentence and the document. We propose new metrics of relevance and redundancy using pointwise mutual information (PMI) between sentences, which can be easily computed by a pre-trained language model. Intuitively, a relevant sentence allows readers to infer the document content (high PMI with the document), and a redundant sentence can be inferred from the summary (high PMI with the summary). We then develop a greedy sentence selection algorithm to maximize relevance and minimize redundancy of extracted sentences. We show that our method outperforms similarity-based methods on datasets in a range of domains including news, medical journal articles, and personal anecdotes.
It is crucial to provide an inter-sentence context in Neural Machine Translation (NMT) models for higher-quality translation. With the aim of using a simple approach to incorporate inter-sentence information, we propose mini-batch embedding (MBE) as a way to represent the features of sentences in a mini-batch. We construct a mini-batch by choosing sentences from the same document, and thus the MBE is expected to have contextual information across sentences. Here, we incorporate MBE in an NMT model, and our experiments show that the proposed method consistently outperforms the translation capabilities of strong baselines and improves writing style or terminology to fit the document’s context.
We investigate how Multilingual BERT (mBERT) encodes grammar by examining how the high-order grammatical feature of morphosyntactic alignment (how different languages define what counts as a “subject”) is manifested across the embedding spaces of different languages. To understand if and how morphosyntactic alignment affects contextual embedding spaces, we train classifiers to recover the subjecthood of mBERT embeddings in transitive sentences (which do not contain overt information about morphosyntactic alignment) and then evaluate them zero-shot on intransitive sentences (where subjecthood classification depends on alignment), within and across languages. We find that the resulting classifier distributions reflect the morphosyntactic alignment of their training languages. Our results demonstrate that mBERT representations are influenced by high-level grammatical features that are not manifested in any one input sentence, and that this is robust across languages. Further examining the characteristics that our classifiers rely on, we find that features such as passive voice, animacy and case strongly correlate with classification decisions, suggesting that mBERT does not encode subjecthood purely syntactically, but that subjecthood embedding is continuous and dependent on semantic and discourse factors, as is proposed in much of the functional linguistics literature. Together, these results provide insight into how grammatical features manifest in contextual embedding spaces, at a level of abstraction not covered by previous work.
Using end-to-end models for speech translation (ST) has increasingly been the focus of the ST community. These models condense the previously cascaded systems by directly converting sound waves into translated text. However, cascaded models have the advantage of including automatic speech recognition output, useful for a variety of practical ST systems that often display transcripts to the user alongside the translations. To bridge this gap, recent work has shown initial progress into the feasibility for end-to-end models to produce both of these outputs. However, all previous work has only looked at this problem from the consecutive perspective, leaving uncertainty on whether these approaches are effective in the more challenging streaming setting. We develop an end-to-end streaming ST model based on a re-translation approach and compare against standard cascading approaches. We also introduce a novel inference method for the joint case, interleaving both transcript and translation in generation and removing the need to use separate decoders. Our evaluation across a range of metrics capturing accuracy, latency, and consistency shows that our end-to-end models are statistically similar to cascading models, while having half the number of parameters. We also find that both systems provide strong translation quality at low latency, keeping 99% of consecutive quality at a lag of just under a second.
This paper explores learning rich self-supervised entity representations from large amounts of associated text. Once pre-trained, these models become applicable to multiple entity-centric tasks such as ranked retrieval, knowledge base completion, question answering, and more. Unlike other methods that harvest self-supervision signals based merely on a local context within a sentence, we radically expand the notion of context to include any available text related to an entity. This enables a new class of powerful, high-capacity representations that can ultimately distill much of the useful information about an entity from multiple text sources, without any human supervision. We present several training strategies that, unlike prior approaches, learn to jointly predict words and entities – strategies we compare experimentally on downstream tasks in the TV-Movies domain, such as MovieLens tag prediction from user reviews and natural language movie search. As evidenced by results, our models match or outperform competitive baselines, sometimes with little or no fine-tuning, and are also able to scale to very large corpora. Finally, we make our datasets and pre-trained models publicly available. This includes Reviews2Movielens, mapping the ~1B word corpus of Amazon movie reviews (He and McAuley, 2016) to MovieLens tags (Harper and Konstan, 2016), as well as Reddit Movie Suggestions with natural language queries and corresponding community recommendations.
Evidence plays a crucial role in any biomedical research narrative, providing justification for some claims and refutation for others. We seek to build models of scientific argument using information extraction methods from full-text papers. We present the capability of automatically extracting text fragments from primary research papers that describe the evidence presented in that paper’s figures, which arguably provides the raw material of any scientific argument made within the paper. We apply richly contextualized deep representation learning pre-trained on biomedical domain corpus to the analysis of scientific discourse structures and the extraction of “evidence fragments” (i.e., the text in the results section describing data presented in a specified subfigure) from a set of biomedical experimental research articles. We first demonstrate our state-of-the-art scientific discourse tagger on two scientific discourse tagging datasets and its transferability to new datasets. We then show the benefit of leveraging scientific discourse tags for downstream tasks such as claim-extraction and evidence fragment detection. Our work demonstrates the potential of using evidence fragments derived from figure spans for improving the quality of scientific claims by cataloging, indexing and reusing evidence fragments as independent documents.
The performance of natural language generation systems has improved substantially with modern neural networks. At test time they typically employ beam search to avoid locally optimal but globally suboptimal predictions. However, due to model errors, a larger beam size can lead to deteriorating performance according to the evaluation metric. For this reason, it is common to rerank the output of beam search, but this relies on beam search to produce a good set of hypotheses, which limits the potential gains. Other alternatives to beam search require changes to the training of the model, which restricts their applicability compared to beam search. This paper proposes incremental beam manipulation, i.e. reranking the hypotheses in the beam during decoding instead of only at the end. This way, hypotheses that are unlikely to lead to a good final output are discarded, and in their place hypotheses that would have been ignored will be considered instead. Applying incremental beam manipulation leads to an improvement of 1.93 and 5.82 BLEU points over vanilla beam search for the test sets of the E2E and WebNLG challenges respectively. The proposed method also outperformed a strong reranker by 1.04 BLEU points on the E2E challenge, while being on par with it on the WebNLG dataset.
Abstractive text summarization aims at compressing the information of a long source document into a rephrased, condensed summary. Despite advances in modeling techniques, abstractive summarization models still suffer from several key challenges: (i) layout bias: they overfit to the style of training corpora; (ii) limited abstractiveness: they are optimized to copying n-grams from the source rather than generating novel abstractive summaries; (iii) lack of transparency: they are not interpretable. In this work, we propose a framework based on document-level structure induction for summarization to address these challenges. To this end, we propose incorporating latent and explicit dependencies across sentences in the source document into end-to-end single-document summarization models. Our framework complements standard encoder-decoder summarization models by augmenting them with rich structure-aware document representations based on implicitly learned (latent) structures and externally-derived linguistic (explicit) structures. We show that our summarization framework, trained on the CNN/DM dataset, improves the coverage of content in the source documents, generates more abstractive summaries by generating more novel n-grams, and incorporates interpretable sentence-level structures, while performing on par with standard baselines.
This paper describes the first report on cross-lingual transfer for semantic dependency parsing. We present the insight that there are twodifferent kinds of cross-linguality, namely sur-face level and mantic level, and try to cap-ture both kinds of cross-linguality by combin-ing annotation projection and model transferof pre-trained language models. Our exper-iments showed that the performance of our graph-based semantic dependency parser almost achieved the approximated upper bound.
Open Information Extraction (OIE) systems seek to compress the factual propositions of a sentence into a series of n-ary tuples. These tuples are useful for downstream tasks in natural language processing like knowledge base creation, textual entailment, and natural language understanding. However, current OIE datasets are limited in both size and diversity. We introduce a new dataset by converting the QA-SRL 2.0 dataset to a large-scale OIE dataset LSOIE. Our LSOIE dataset is 20 times larger than the next largest human-annotated OIE dataset. We construct and evaluate several benchmark OIE models on LSOIE, providing baselines for future improvements on the task. Our LSOIE data, models, and code are made publicly available.
Large Transformer-based language models can aid human authors by suggesting plausible continuations of text written so far. However, current interactive writing assistants do not allow authors to guide text generation in desired topical directions. To address this limitation, we design a framework that displays multiple candidate upcoming topics, of which a user can select a subset to guide the generation. Our framework consists of two components: (1) a method that produces a set of candidate topics by predicting the centers of word clusters in the possible continuations, and (2) a text generation model whose output adheres to the chosen topics. The training of both components is self-supervised, using only unlabeled text. Our experiments demonstrate that our topic options are better than those of standard clustering approaches, and our framework often generates fluent sentences related to the chosen topics, as judged by automated metrics and crowdsourced workers.
Abstractive summarization systems generally rely on large collections of document-summary pairs. However, the performance of abstractive systems remains a challenge due to the unavailability of the parallel data for low-resource languages like Bengali. To overcome this problem, we propose a graph-based unsupervised abstractive summarization system in the single-document setting for Bengali text documents, which requires only a Part-Of-Speech (POS) tagger and a pre-trained language model trained on Bengali texts. We also provide a human-annotated dataset with document-summary pairs to evaluate our abstractive model and to support the comparison of future abstractive summarization systems of the Bengali Language. We conduct experiments on this dataset and compare our system with several well-established unsupervised extractive summarization systems. Our unsupervised abstractive summarization model outperforms the baselines without being exposed to any human-annotated reference summaries.
The ability to quantify incivility online, in news and in congressional debates is of great interest to political scientists. Computational tools for detecting online incivility for English are now fairly accessible and potentially could be applied more broadly. We test the Jigsaw Perspective API for its ability to detect the degree of incivility on a corpus that we developed, consisting of manual annotations of civility in American news. We demonstrate that toxicity models, as exemplified by Perspective, are inadequate for the analysis of incivility in news. We carry out error analysis that points to the need to develop methods to remove spurious correlations between words often mentioned in the news, especially identity descriptors and incivility. Without such improvements, applying Perspective or similar models on news is likely to lead to wrong conclusions, that are not aligned with the human perception of incivility.
This paper presents a finite-state computational model of the verbal morphology of Michif. Michif, the official language of the Métis peoples, is a uniquely mixed language with Algonquian and French origins. It is spoken across the Métis homelands in what is now called Canada and the United States, but it is highly endangered with less than 100 speakers. The verbal morphology is remarkably complex, as the already polysynthetic Algonquian patterns are combined with French elements and unique morpho-phonological interactions.The model presented in this paper, LI VERB KAA-OOSHITAHK DI MICHIF handles this complexity by using a series of composed finite-state transducers to model the concatenative morphology and phonological rule alternations that are unique to Michif. Such a rule-based approach is necessary as there is insufficient language data for an approach that uses machine learning. A language model such as LI VERB KAA-OOSHITAHK DI MICHIF furthers the goals of Indigenous computational linguistics in Canada while also supporting the creation of tools for documentation, education, and revitalization that are desired by the Métis community.
User stance detection entails ascertaining the position of a user towards a target, such as an entity, topic, or claim. Recent work that employs unsupervised classification has shown that performing stance detection on vocal Twitter users, who have many tweets on a target, can be highly accurate (+98%). However, such methods perform poorly or fail completely for less vocal users, who may have authored only a few tweets about a target. In this paper, we tackle stance detection for such users using two approaches. In the first approach, we improve user-level stance detection by representing tweets using contextualized embeddings, which capture latent meanings of words in context. We show that this approach outperforms two strong baselines and achieves 89.6% accuracy and 91.3% macro F-measure on eight controversial topics. In the second approach, we expand the tweets of a given user using their Twitter timeline tweets, which may not be topically relevant, and then we perform unsupervised classification of the user, which entails clustering a user with other users in the training set. This approach achieves 95.6% accuracy and 93.1% macro F-measure.
Much recent work suggests that incorporating syntax information from dependency trees can improve task-specific transformer models. However, the effect of incorporating dependency tree information into pre-trained transformer models (e.g., BERT) remains unclear, especially given recent studies highlighting how these models implicitly encode syntax. In this work, we systematically study the utility of incorporating dependency trees into pre-trained transformers on three representative information extraction tasks: semantic role labeling (SRL), named entity recognition, and relation extraction. We propose and investigate two distinct strategies for incorporating dependency structure: a late fusion approach, which applies a graph neural network on the output of a transformer, and a joint fusion approach, which infuses syntax structure into the transformer attention layers. These strategies are representative of prior work, but we introduce additional model design elements that are necessary for obtaining improved performance. Our empirical analysis demonstrates that these syntax-infused transformers obtain state-of-the-art results on SRL and relation extraction tasks. However, our analysis also reveals a critical shortcoming of these models: we find that their performance gains are highly contingent on the availability of human-annotated dependency parses, which raises important questions regarding the viability of syntax-augmented transformers in real-world applications.
Opinion summarization is the task of automatically generating summaries for a set of reviews about a specific target (e.g., a movie or a product). Since the number of reviews for each target can be prohibitively large, neural network-based methods follow a two-stage approach where an extractive step first pre-selects a subset of salient opinions and an abstractive step creates the summary while conditioning on the extracted subset. However, the extractive model leads to loss of information which may be useful depending on user needs. In this paper we propose a summarization framework that eliminates the need to rely only on pre-selected content and waste possibly useful information, especially when customizing summaries. The framework enables the use of all input reviews by first condensing them into multiple dense vectors which serve as input to an abstractive model. We showcase an effective instantiation of our framework which produces more informative summaries and also allows to take user preferences into account using our zero-shot customization technique. Experimental results demonstrate that our model improves the state of the art on the Rotten Tomatoes dataset and generates customized summaries effectively.
In contrast to their word- or sentence-level counterparts, character embeddings are still poorly understood. We aim at closing this gap with an in-depth study of English character embeddings. For this, we use resources from research on grapheme–color synesthesia – a neuropsychological phenomenon where letters are associated with colors –, which give us insight into which characters are similar for synesthetes and how characters are organized in color space. Comparing 10 different character embeddings, we ask: How similar are character embeddings to a synesthete’s perception of characters? And how similar are character embeddings extracted from different models? We find that LSTMs agree with humans more than transformers. Comparing across tasks, grapheme-to-phoneme conversion results in the most human-like character embeddings. Finally, ELMo embeddings differ from both humans and other models.
Standard evaluations of Grammatical Error Correction (GEC) systems make use of a fixed reference text generated relative to the original text; they show, even when using multiple references, that we have a long way to go. This analysis paper studies the performance of GEC systems relative to closest-gold – a gold reference text created relative to the output of a system. Surprisingly, we show that the real performance is 20-40 points better than standard evaluations show. Moreover, the performance remains high even when considering any of the top-10 hypotheses produced by a system. Importantly, the type of mistakes corrected by lower-ranked hypotheses differs in interesting ways from the top one, providing an opportunity to focus on a range of errors – local spelling and grammar edits vs. more complex lexical improvements. Our study shows these results in English and Russian, and thus provides a preliminary proposal for a more realistic evaluation of GEC systems.
Spotting a lie is challenging but has an enormous potential impact on security as well as private and public safety. Several NLP methods have been proposed to classify texts as truthful or deceptive. In most cases, however, the target texts’ preceding context is not considered. This is a severe limitation, as any communication takes place in context, not in a vacuum, and context can help to detect deception. We study a corpus of Italian dialogues containing deceptive statements and implement deep neural models that incorporate various linguistic contexts. We establish a new state-of-the-art identifying deception and find that not all context is equally useful to the task. Only the texts closest to the target, if from the same speaker (rather than questions by an interlocutor), boost performance. We also find that the semantic information in language models such as BERT contributes to the performance. However, BERT alone does not capture the implicit knowledge of deception cues: its contribution is conditional on the concurrent use of attention to learn cues from BERT’s representations.
We present a novel approach to efficiently learn a simultaneous translation model with coupled programmer-interpreter policies. First, we present an algorithmic oracle to produce oracle READ/WRITE actions for training bilingual sentence-pairs using the notion of word alignments. This oracle actions are designed to capture enough information from the partial input before writing the output. Next, we perform a coupled scheduled sampling to effectively mitigate the exposure bias when learning both policies jointly with imitation learning. Experiments on six language-pairs show our method outperforms strong baselines in terms of translation quality quality while keeping the delay low.
This paper proposes a new problem of complementary evidence identification for open-domain question answering (QA). The problem aims to efficiently find a small set of passages that covers full evidence from multiple aspects as to answer a complex question. To this end, we proposes a method that learns vector representations of passages and models the sufficiency and diversity within the selected set, in addition to the relevance between the question and passages. Our experiments demonstrate that our method considers the dependence within the supporting evidence and significantly improves the accuracy of complementary evidence selection in QA domain.
A key challenge for abstractive summarization is ensuring factual consistency of the generated summary with respect to the original document. For example, state-of-the-art models trained on existing datasets exhibit entity hallucination, generating names of entities that are not present in the source document. We propose a set of new metrics to quantify the entity-level factual consistency of generated summaries and we show that the entity hallucination problem can be alleviated by simply filtering the training data. In addition, we propose a summary-worthy entity classification task to the training process as well as a joint entity and summary generation approach, which yield further improvements in entity level metrics.
Despite improvements in performances on different natural language generation tasks, deep neural models are prone to hallucinating facts that are incorrect or nonexistent. Different hypotheses are proposed and examined separately for different tasks, but no systematic explanations are available across these tasks. In this study, we draw connections between hallucinations and predictive uncertainty in conditional language generation. We investigate their relationship in both image captioning and data-to-text generation and propose a simple extension to beam search to reduce hallucination. Our analysis shows that higher predictive uncertainty corresponds to a higher chance of hallucination. Epistemic uncertainty is more indicative of hallucination than aleatoric or total uncertainties. It helps to achieve better results of trading performance in standard metric for less hallucination with the proposed beam search variant.
Most of the previous work on Event Detection (ED) has only considered the datasets with a small number of event types (i.e., up to 38 types). In this work, we present the first study on fine-grained ED (FED) where the evaluation dataset involves much more fine-grained event types (i.e., 449 types). We propose a novel method to transform the Semcor dataset for Word Sense Disambiguation into a large and high-quality dataset for FED. Extensive evaluation of the current ED methods is conducted to demonstrate the challenges of the generated datasets for FED, calling for more research effort in this area.
Pretrained language models like BERT have achieved good results on NLP tasks, but are impractical on resource-limited devices due to memory footprint. A large fraction of this footprint comes from the input embeddings with large input vocabulary and embedding dimensions. Existing knowledge distillation methods used for model compression cannot be directly applied to train student models with reduced vocabulary sizes. To this end, we propose a distillation method to align the teacher and student embeddings via mixed-vocabulary training. Our method compresses BERT-LARGE to a task-agnostic model with smaller vocabulary and hidden dimensions, which is an order of magnitude smaller than other distilled BERT models and offers a better size-accuracy trade-off on language understanding benchmarks as well as a practical dialogue task.
Adversarial learning can learn fairer and less biased models of language processing than standard training. However, current adversarial techniques only partially mitigate the problem of model bias, added to which their training procedures are often unstable. In this paper, we propose a novel approach to adversarial learning based on the use of multiple diverse discriminators, whereby discriminators are encouraged to learn orthogonal hidden representations from one another. Experimental results show that our method substantially improves over standard adversarial removal methods, in terms of reducing bias and stability of training.
GQA (CITATION) is a dataset for real-world visual reasoning and compositional question answering. We found that many answers predicted by the best vision-language models on the GQA dataset do not match the ground-truth answer but still are semantically meaningful and correct in the given context. In fact, this is the case with most existing visual question answering (VQA) datasets where they assume only one ground-truth answer for each question. We propose Alternative Answer Sets (AAS) of ground-truth answers to address this limitation, which is created automatically using off-the-shelf NLP tools. We introduce a semantic metric based on AAS and modify top VQA solvers to support multiple plausible answers for a question. We implement this approach on the GQA dataset and show the performance improvements.
Adding linguistic information (syntax or semantics) to neural machine translation (NMT) have mostly focused on using point estimates from pre-trained models. Directly using the capacity of massive pre-trained contextual word embedding models such as BERT(Devlin et al., 2019) has been marginally useful in NMT because effective fine-tuning is difficult to obtain for NMT without making training brittle and unreliable. We augment NMT by extracting dense fine-tuned vector-based linguistic information from BERT instead of using point estimates. Experimental results show that our method of incorporating linguistic information helps NMT to generalize better in a variety of training contexts and is no more difficult to train than conventional Transformer-based NMT.
Linguistically informed analyses of language models (LMs) contribute to the understanding and improvement of such models. Here, we introduce the corpus of Chinese linguistic minimal pairs (CLiMP) to investigate what knowledge Chinese LMs acquire. CLiMP consists of sets of 1000 minimal pairs (MPs) for 16 syntactic contrasts in Chinese, covering 9 major Chinese linguistic phenomena. The MPs are semi-automatically generated, and human agreement with the labels in CLiMP is 95.8%. We evaluate 11 different LMs on CLiMP, covering n-grams, LSTMs, and Chinese BERT. We find that classifier–noun agreement and verb complement selection are the phenomena that models generally perform best at. However, models struggle the most with the ba construction, binding, and filler-gap dependencies. Overall, Chinese BERT achieves an 81.8% average accuracy, while the performances of LSTMs and 5-grams are only moderately above chance level.
While the attention heatmaps produced by neural machine translation (NMT) models seem insightful, there is little evidence that they reflect a model’s true internal reasoning. We provide a measure of faithfulness for NMT based on a variety of stress tests where attention weights which are crucial for prediction are perturbed and the model should alter its predictions if the learned weights are a faithful explanation of the predictions. We show that our proposed faithfulness measure for NMT models can be improved using a novel differentiable objective that rewards faithful behaviour by the model through probability divergence. Our experimental results on multiple language pairs show that our objective function is effective in increasing faithfulness and can lead to a useful analysis of NMT model behaviour and more trustworthy attention heatmaps. Our proposed objective improves faithfulness without reducing the translation quality and has a useful regularization effect on the NMT model and can even improve translation quality in some cases.
Commonly used information retrieval methods such as TF-IDF in open-domain question answering (QA) systems are insufficient to capture deep semantic matching that goes beyond lexical overlaps. Some recent studies consider the retrieval process as maximum inner product search (MIPS) using dense question and paragraph representations, achieving promising results on several information-seeking QA datasets. However, the pretraining of the dense vector representations is highly resource-demanding, e.g., requires a very large batch size and lots of training steps. In this work, we propose a sample-efficient method to pretrain the paragraph encoder. First, instead of using heuristically created pseudo question-paragraph pairs for pretraining, we use an existing pretrained sequence-to-sequence model to build a strong question generator that creates high-quality pretraining data. Second, we propose a simple progressive pretraining algorithm to ensure the existence of effective negative samples in each batch. Across three open-domain QA datasets, our method consistently outperforms a strong dense retrieval baseline that uses 6 times more computation for training. On two of the datasets, our method achieves more than 4-point absolute improvement in terms of answer exact match.
Real-world knowledge graphs are often characterized by low-frequency relations—a challenge that has prompted an increasing interest in few-shot link prediction methods. These methods perform link prediction for a set of new relations, unseen during training, given only a few example facts of each relation at test time. In this work, we perform a systematic study on a spectrum of models derived by generalizing the current state of the art for few-shot link prediction, with the goal of probing the limits of learning in this few-shot setting. We find that a simple, zero-shot baseline — which ignores any relation-specific information — achieves surprisingly strong performance. Moreover, experiments on carefully crafted synthetic datasets show that having only a few examples of a relation fundamentally limits models from using fine-grained structural information and only allows for exploiting the coarse-grained positional information of entities. Together, our findings challenge the implicit assumptions and inductive biases of prior work and highlight new directions for research in this area.
At the heart of text based neural models lay word representations, which are powerful but occupy a lot of memory making it challenging to deploy to devices with memory constraints such as mobile phones, watches and IoT. To surmount these challenges, we introduce ProFormer – a projection based transformer architecture that is faster and lighter making it suitable to deploy to memory constraint devices and preserve user privacy. We use LSH projection layer to dynamically generate word representations on-the-fly without embedding lookup tables leading to significant memory footprint reduction from O(V.d) to O(T), where V is the vocabulary size, d is the embedding dimension size and T is the dimension of the LSH projection representation.We also propose a local projection attention (LPA) layer, which uses self-attention to transform the input sequence of N LSH word projections into a sequence of N/K representations reducing the computations quadratically by O(Kˆ2). We evaluate ProFormer on multiple text classification tasks and observed improvements over prior state-of-the-art on-device approaches for short text classification and comparable performance for long text classification tasks. ProFormer is also competitive with other popular but highly resource-intensive approaches like BERT and even outperforms small-sized BERT variants with significant resource savings – reduces the embedding memory footprint from 92.16 MB to 1.7 KB and requires 16x less computation overhead, which is very impressive making it the fastest and smallest on-device model.
We consider the problem of multi-label classification, where the labels lie on a hierarchy. However, unlike most existing works in hierarchical multi-label classification, we do not assume that the label-hierarchy is known. Encouraged by the recent success of hyperbolic embeddings in capturing hierarchical relations, we propose to jointly learn the classifier parameters as well as the label embeddings. Such a joint learning is expected to provide a twofold advantage: i) the classifier generalises better as it leverages the prior knowledge of existence of a hierarchy over the labels, and ii) in addition to the label co-occurrence information, the label-embedding may benefit from the manifold structure of the input datapoints, leading to embeddings that are more faithful to the label hierarchy. We propose a novel formulation for the joint learning and empirically evaluate its efficacy. The results show that the joint learning improves over the baseline that employs label co-occurrence based pre-trained hyperbolic embeddings. Moreover, the proposed classifiers achieve state-of-the-art generalization on standard benchmarks. We also present evaluation of the hyperbolic embeddings obtained by joint learning and show that they represent the hierarchy more accurately than the other alternatives.
Typical ASR systems segment the input audio into utterances using purely acoustic information, which may not resemble the sentence-like units that are expected by conventional machine translation (MT) systems for Spoken Language Translation. In this work, we propose a model for correcting the acoustic segmentation of ASR models for low-resource languages to improve performance on downstream tasks. We propose the use of subtitles as a proxy dataset for correcting ASR acoustic segmentation, creating synthetic acoustic utterances by modeling common error modes. We train a neural tagging model for correcting ASR acoustic segmentation and show that it improves downstream performance on MT and audio-document cross-language information retrieval (CLIR).
By supporting multi-modal retrieval training and evaluation, image captioning datasets have spurred remarkable progress on representation learning. Unfortunately, datasets have limited cross-modal associations: images are not paired with other images, captions are only paired with other captions of the same image, there are no negative associations and there are missing positive cross-modal associations. This undermines research into how inter-modality learning impacts intra-modality tasks. We address this gap with Crisscrossed Captions (CxC), an extension of the MS-COCO dataset with human semantic similarity judgments for 267,095 intra- and inter-modality pairs. We report baseline results on CxC for strong existing unimodal and multimodal models. We also evaluate a multitask dual encoder trained on both image-caption and caption-caption pairs that crucially demonstrates CxC’s value for measuring the influence of intra- and inter-modality learning.
Recently, there has been a strong interest in developing natural language applications that live on personal devices such as mobile phones, watches and IoT with the objective to preserve user privacy and have low memory. Advances in Locality-Sensitive Hashing (LSH)-based projection networks have demonstrated state-of-the-art performance in various classification tasks without explicit word (or word-piece) embedding lookup tables by computing on-the-fly text representations. In this paper, we show that the projection based neural classifiers are inherently robust to misspellings and perturbations of the input text. We empirically demonstrate that the LSH projection based classifiers are more robust to common misspellings compared to BiLSTMs (with both word-piece & word-only tokenization) and fine-tuned BERT based methods. When subject to misspelling attacks, LSH projection based classifiers had a small average accuracy drop of 2.94% across multiple classifications tasks, while the fine-tuned BERT model accuracy had a significant drop of 11.44%.
Current state-of-the-art systems for joint entity relation extraction (Luan et al., 2019; Wad-den et al., 2019) usually adopt the multi-task learning framework. However, annotations for these additional tasks such as coreference resolution and event extraction are always equally hard (or even harder) to obtain. In this work, we propose a pre-training method ENPAR to improve the joint extraction performance. ENPAR requires only the additional entity annotations that are much easier to collect. Unlike most existing works that only consider incorporating entity information into the sentence encoder, we further utilize the entity pair information. Specifically, we devise four novel objectives,i.e., masked entity typing, masked entity prediction, adversarial context discrimination, and permutation prediction, to pre-train an entity encoder and an entity pair encoder. Comprehensive experiments show that the proposed pre-training method achieves significant improvement over BERT on ACE05, SciERC, and NYT, and outperforms current state-of-the-art on ACE05.
Traditional data augmentation aims to increase the coverage of the input distribution by generating augmented examples that strongly resemble original samples in an online fashion where augmented examples dominate training. In this paper, we propose an alternative perspective—a multi-task view (MTV) of data augmentation—in which the primary task trains on original examples and the auxiliary task trains on augmented examples. In MTV data augmentation, both original and augmented samples are weighted substantively during training, relaxing the constraint that augmented examples must resemble original data and thereby allowing us to apply stronger augmentation functions. In empirical experiments using four common data augmentation techniques on three benchmark text classification datasets, we find that using the MTV leads to higher and more robust performance than traditional augmentation.
Tables in web documents are pervasive and can be directly used to answer many of the queries searched on the web, motivating their integration in question answering. Very often information presented in tables is succinct and hard to interpret with standard language representations. On the other hand, tables often appear within textual context, such as an article describing the table. Using the information from an article as additional context can potentially enrich table representations. In this work we aim to improve question answering from tables by refining table representations based on information from surrounding text. We also present an effective method to combine text and table-based predictions for question answering from full documents, obtaining significant improvements on the Natural Questions dataset (Kwiatkowski et al., 2019).
Cross-lingual transfer is a leading technique for parsing low-resource languages in the absence of explicit supervision. Simple ‘direct transfer’ of a learned model based on a multilingual input encoding has provided a strong benchmark. This paper presents a method for unsupervised cross-lingual transfer that improves over direct transfer systems by using their output as implicit supervision as part of self-training on unlabelled text in the target language. The method assumes minimal resources and provides maximal flexibility by (a) accepting any pre-trained arc-factored dependency parser; (b) assuming no access to source language data; (c) supporting both projective and non-projective parsing; and (d) supporting multi-source transfer. With English as the source language, we show significant improvements over state-of-the-art transfer models on both distant and nearby languages, despite our conceptually simpler approach. We provide analyses of the choice of source languages for multi-source transfer, and the advantage of non-projective parsing. Our code is available online.
A recent topic of research in natural language generation has been the development of automatic response generation modules that can automatically respond to a user’s utterance in an empathetic manner. Previous research has tackled this task using neural generative methods by augmenting emotion classes with the input sequences. However, the outputs by these models may be inconsistent. We employ multi-task learning to predict the emotion label and to generate a viable response for a given utterance using a common encoder with multiple decoders. Our proposed encoder-decoder model consists of a self-attention based encoder and a decoder with dot product attention mechanism to generate response with a specified emotion. We use the focal loss to handle imbalanced data distribution, and utilize the consistency loss to allow coherent decoding by the decoders. Human evaluation reveals that our model produces more emotionally pertinent responses. In addition, our model outperforms multiple strong baselines on automatic evaluation measures such as F1 and BLEU scores, thus resulting in more fluent and adequate responses.
Multiple studies have demonstrated that behaviors expressed on online social media platforms can indicate the mental health state of an individual. The widespread availability of such data has spurred interest in mental health research, using several datasets where individuals are labeled with mental health conditions. While previous research has raised concerns about possible biases in models produced from this data, no study has investigated how these biases manifest themselves with regards to demographic groups in data, such as gender and racial/ethnic groups. Here, we analyze the fairness of depression classifiers trained on Twitter data with respect to gender and racial demographic groups. We find that model performance differs for underrepresented groups, and we investigate sources of these biases beyond data representation. Our study results in recommendations on how to avoid these biases in future research.
Scaling semantic parsing models for task-oriented dialog systems to new languages is often expensive and time-consuming due to the lack of available datasets. Available datasets suffer from several shortcomings: a) they contain few languages b) they contain small amounts of labeled examples per language c) they are based on the simple intent and slot detection paradigm for non-compositional queries. In this paper, we present a new multilingual dataset, called MTOP, comprising of 100k annotated utterances in 6 languages across 11 domains. We use this dataset and other publicly available datasets to conduct a comprehensive benchmarking study on using various state-of-the-art multilingual pre-trained models for task-oriented semantic parsing. We achieve an average improvement of +6.3 points on Slot F1 for the two existing multilingual datasets, over best results reported in their experiments. Furthermore, we demonstrate strong zero-shot performance using pre-trained models combined with automatic translation and alignment, and a proposed distant supervision method to reduce the noise in slot label projection.
We tackle the task of adapting event extractors to new domains without labeled data, by aligning the marginal distributions of source and target domains. As a testbed, we create two new event extraction datasets using English texts from two medical domains: (i) clinical notes, and (ii) doctor-patient conversations. We test the efficacy of three marginal alignment techniques: (i) adversarial domain adaptation (ADA), (ii) domain adaptive fine-tuning (DAFT), and (iii) a new instance weighting technique based on language model likelihood scores (LIW). LIW and DAFT improve over a no-transfer BERT baseline on both domains, but ADA only improves on notes. Deeper analysis of performance under different types of shifts (e.g., lexical shift, semantic shift) explains some of the variations among models. Our best-performing models reach F1 scores of 70.0 and 72.9 on notes and conversations respectively, using no labeled target data.
When Question-Answering (QA) systems are deployed in the real world, users query them through a variety of interfaces, such as speaking to voice assistants, typing questions into a search engine, or even translating questions to languages supported by the QA system. While there has been significant community attention devoted to identifying correct answers in passages assuming a perfectly formed question, we show that components in the pipeline that precede an answering engine can introduce varied and considerable sources of error, and performance can degrade substantially based on these upstream noise sources even for powerful pre-trained QA models. We conclude that there is substantial room for progress before QA systems can be effectively deployed, highlight the need for QA evaluation to expand to consider real-world use, and hope that our findings will spur greater community interest in the issues that arise when our systems actually need to be of utility to humans.
Referential games offer a grounded learning environment for neural agents which accounts for the fact that language is functionally used to communicate. However, they do not take into account a second constraint considered to be fundamental for the shape of human language: that it must be learnable by new language learners. Cogswell et al. (2019) introduced cultural transmission within referential games through a changing population of agents to constrain the emerging language to be learnable. However, the resulting languages remain inherently biased by the agents’ underlying capabilities. In this work, we introduce Language Transmission Simulator to model both cultural and architectural evolution in a population of agents. As our core contribution, we empirically show that the optimal situation is to take into account also the learning biases of the language learners and thus let language and agents co-evolve. When we allow the agent population to evolve through architectural evolution, we achieve across the board improvements on all considered metrics and surpass the gains made with cultural transmission. These results stress the importance of studying the underlying agent architecture and pave the way to investigate the co-evolution of language and agent in language emergence studies.
Answer Sentence Selection (AS2) is an efficient approach for the design of open-domain Question Answering (QA) systems. In order to achieve low latency, traditional AS2 models score question-answer pairs individually, ignoring any information from the document each potential answer was extracted from. In contrast, more computationally expensive models designed for machine reading comprehension tasks typically receive one or more passages as input, which often results in better accuracy. In this work, we present an approach to efficiently incorporate contextual information in AS2 models. For each answer candidate, we first use unsupervised similarity techniques to extract relevant sentences from its source document, which we then feed into an efficient transformer architecture fine-tuned for AS2. Our best approach, which leverages a multi-way attention architecture to efficiently encode context, improves 6% to 11% over non-contextual state of the art in AS2 with minimal impact on system latency. All experiments in this work were conducted in English.
Pre-trained language models like BERT achieve superior performances in various NLP tasks without explicit consideration of syntactic information. Meanwhile, syntactic information has been proved to be crucial for the success of NLP applications. However, how to incorporate the syntax trees effectively and efficiently into pre-trained Transformers is still unsettled. In this paper, we address this problem by proposing a novel framework named Syntax-BERT. This framework works in a plug-and-play mode and is applicable to an arbitrary pre-trained checkpoint based on Transformer architecture. Experiments on various datasets of natural language understanding verify the effectiveness of syntax trees and achieve consistent improvement over multiple pre-trained models, including BERT, RoBERTa, and T5.
Neural networks (NN) applied to natural language processing (NLP) are becoming deeper and more complex, making them increasingly difficult to understand and interpret. Even in applications of limited scope on fixed data, the creation of these complex “black-boxes” creates substantial challenges for debugging, understanding, and generalization. But rapid development in this field has now lead to building more straightforward and interpretable models. We propose a new technique (DISK-CSV) to distill knowledge concurrently from any neural network architecture for text classification, captured as a lightweight interpretable/explainable classifier. Across multiple datasets, our approach achieves better performance than the target black-box. In addition, our approach provides better explanations than existing techniques.
Since the popularization of the Transformer as a general-purpose feature encoder for NLP, many studies have attempted to decode linguistic structure from its novel multi-head attention mechanism. However, much of such work focused almost exclusively on English — a language with rigid word order and a lack of inflectional morphology. In this study, we present decoding experiments for multilingual BERT across 18 languages in order to test the generalizability of the claim that dependency syntax is reflected in attention patterns. We show that full trees can be decoded above baseline accuracy from single attention heads, and that individual relations are often tracked by the same heads across languages. Furthermore, in an attempt to address recent debates about the status of attention as an explanatory mechanism, we experiment with fine-tuning mBERT on a supervised parsing objective while freezing different series of parameters. Interestingly, in steering the objective to learn explicit linguistic structure, we find much of the same structure represented in the resulting attention patterns, with interesting differences with respect to which parameters are frozen.
Modeling the relations between text spans in a document is a crucial yet challenging problem for extractive summarization. Various kinds of relations exist among text spans of different granularity, such as discourse relations between elementary discourse units and coreference relations between phrase mentions. In this paper, we propose a heterogeneous graph based model for extractive summarization that incorporates both discourse and coreference relations. The heterogeneous graph contains three types of nodes, each corresponds to text spans of different granularity. Experimental results on a benchmark summarization dataset verify the effectiveness of our proposed method.
We introduce a Content-based Document Alignment approach (CDA), an efficient method to align multilingual web documents based on content in creating parallel training data for machine translation (MT) systems operating at the industrial level. CDA works in two steps: (i) projecting documents of a web domain to a shared multilingual space; then (ii) aligning them based on the similarity of their representations in such space. We leverage lexical translation models to build vector representations using TF×IDF. CDA achieves performance comparable with state-of-the-art systems in the WMT-16 Bilingual Document Alignment Shared Task benchmark while operating in multilingual space. Besides, we created two web-scale datasets to examine the robustness of CDA in an industrial setting involving up to 28 languages and millions of documents. The experiments show that CDA is robust, cost-effective, and is significantly superior in (i) processing large and noisy web data and (ii) scaling to new and low-resourced languages.
Numerical tables are widely used to present experimental results in scientific papers. For table understanding, a metric-type is essential to discriminate numbers in the tables. We introduce a new information extraction task, metric-type identification from multi-level header numerical tables, and provide a dataset extracted from scientific papers consisting of header tables, captions, and metric-types. We then propose two joint-learning neural classification and generation schemes featuring pointer-generator-based and BERT-based models. Our results show that the joint models can handle both in-header and out-of-header metric-type identification problems.
Affect preferences vary with user demographics, and tapping into demographic information provides important cues about the users’ language preferences. In this paper, we utilize the user demographics and propose EmpathBERT, a demographic-aware framework for empathy prediction based on BERT. Through several comparative experiments, we show that EmpathBERT surpasses traditional machine learning and deep learning models, and illustrate the importance of user demographics, for predicting empathy and distress in user responses to stimulative news articles. We also highlight the importance of affect information in the responses by developing affect-aware models to predict user demographic attributes.
We evaluate the ability of Bert embeddings to represent tense information, taking French and Chinese as a case study. In French, the tense information is expressed by verb morphology and can be captured by simple surface information. On the contrary, tense interpretation in Chinese is driven by abstract, lexical, syntactic and even pragmatic information. We show that while French tenses can easily be predicted from sentence representations, results drop sharply for Chinese, which suggests that Bert is more likely to memorize shallow patterns from the training data rather than uncover abstract properties.
Traditional NLP has long held (supervised) syntactic parsing necessary for successful higher-level semantic language understanding (LU). The recent advent of end-to-end neural models, self-supervised via language modeling (LM), and their success on a wide range of LU tasks, however, questions this belief. In this work, we empirically investigate the usefulness of supervised parsing for semantic LU in the context of LM-pretrained transformer networks. Relying on the established fine-tuning paradigm, we first couple a pretrained transformer with a biaffine parsing head, aiming to infuse explicit syntactic knowledge from Universal Dependencies treebanks into the transformer. We then fine-tune the model for LU tasks and measure the effect of the intermediate parsing training (IPT) on downstream LU task performance. Results from both monolingual English and zero-shot language transfer experiments (with intermediate target-language parsing) show that explicit formalized syntax, injected into transformers through IPT, has very limited and inconsistent effect on downstream LU performance. Our results, coupled with our analysis of transformers’ representation spaces before and after intermediate parsing, make a significant step towards providing answers to an essential question: how (un)availing is supervised parsing for high-level semantic natural language understanding in the era of large neural models?
Most of the recent work on terminology integration in machine translation has assumed that terminology translations are given already inflected in forms that are suitable for the target language sentence. In day-to-day work of professional translators, however, it is seldom the case as translators work with bilingual glossaries where terms are given in their dictionary forms; finding the right target language form is part of the translation process. We argue that the requirement for apriori specified target language forms is unrealistic and impedes the practical applicability of previous work. In this work, we propose to train machine translation systems using a source-side data augmentation method that annotates randomly selected source language words with their target language lemmas. We show that systems trained on such augmented data are readily usable for terminology integration in real-life translation scenarios. Our experiments on terminology translation into the morphologically complex Baltic and Uralic languages show an improvement of up to 7 BLEU points over baseline systems with no means for terminology integration and an average improvement of 4 BLEU points over the previous work. Results of the human evaluation indicate a 47.7% absolute improvement over the previous work in term translation accuracy when translating into Latvian.
We propose a novel hybrid approach to lemmatization that enhances the seq2seq neural model with additional lemmas extracted from an external lexicon or a rule-based system. During training, the enhanced lemmatizer learns both to generate lemmas via a sequential decoder and copy the lemma characters from the external candidates supplied during run-time. Our lemmatizer enhanced with candidates extracted from the Apertium morphological analyzer achieves statistically significant improvements compared to baseline models not utilizing additional lemma information, achieves an average accuracy of 97.25% on a set of 23 UD languages, which is 0.55% higher than obtained with the Stanford Stanza model on the same set of languages. We also compare with other methods of integrating external data into lemmatization and show that our enhanced system performs considerably better than a simple lexicon extension method based on the Stanza system, and it achieves complementary improvements w.r.t. the data augmentation method.
We introduce the task of historical text summarisation, where documents in historical forms of a language are summarised in the corresponding modern language. This is a fundamentally important routine to historians and digital humanities researchers but has never been automated. We compile a high-quality gold-standard text summarisation dataset, which consists of historical German and Chinese news from hundreds of years ago summarised in modern German or Chinese. Based on cross-lingual transfer learning techniques, we propose a summarisation model that can be trained even with no cross-lingual (historical to modern) parallel data, and further benchmark it against state-of-the-art algorithms. We report automatic and human evaluations that distinguish the historic to modern language summarisation task from standard cross-lingual summarisation (i.e., modern to modern language), highlight the distinctness and value of our dataset, and demonstrate that our transfer learning approach outperforms standard cross-lingual benchmarks on this task.
Biased associations have been a challenge in the development of classifiers for detecting toxic language, hindering both fairness and accuracy. As potential solutions, we investigate recently introduced debiasing methods for text classification datasets and models, as applied to toxic language detection. Our focus is on lexical (e.g., swear words, slurs, identity mentions) and dialectal markers (specifically African American English). Our comprehensive experiments establish that existing methods are limited in their ability to prevent biased behavior in current toxicity detectors. We then propose an automatic, dialect-aware data correction method, as a proof-of-concept. Despite the use of synthetic labels, this method reduces dialectal associations with toxicity. Overall, our findings show that debiasing a model trained on biased toxic language data is not as effective as simply relabeling the data to remove existing biases.
Effective fusion of data from multiple modalities, such as video, speech, and text, is challenging due to the heterogeneous nature of multimodal data. In this paper, we propose adaptive fusion techniques that aim to model context from different modalities effectively. Instead of defining a deterministic fusion operation, such as concatenation, for the network, we let the network decide “how” to combine a given set of multimodal features more effectively. We propose two networks: 1) Auto-Fusion, which learns to compress information from different modalities while preserving the context, and 2) GAN-Fusion, which regularizes the learned latent space given context from complementing modalities. A quantitative evaluation on the tasks of multimodal machine translation and emotion recognition suggests that our lightweight, adaptive networks can better model context from other modalities than existing methods, many of which employ massive transformer-based networks.
This paper introduces the novel task of scene segmentation on narrative texts and provides an annotated corpus, a discussion of the linguistic and narrative properties of the task and baseline experiments towards automatic solutions. A scene here is a segment of the text where time and discourse time are more or less equal, the narration focuses on one action and location and character constellations stay the same. The corpus we describe consists of German-language dime novels (550k tokens) that have been annotated in parallel, achieving an inter-annotator agreement of gamma = 0.7. Baseline experiments using BERT achieve an F1 score of 24%, showing that the task is very challenging. An automatic scene segmentation paves the way towards processing longer narrative texts like tales or novels by breaking them down into smaller, coherent and meaningful parts, which is an important stepping stone towards the reconstruction of plot in Computational Literary Studies but also can serve to improve tasks like coreference resolution.
The conceptualization of a claim lies at the core of argument mining. The segregation of claims is complex, owing to the divergence in textual syntax and context across different distributions. Another pressing issue is the unavailability of labeled unstructured text for experimentation. In this paper, we propose LESA, a framework which aims at advancing headfirst into expunging the former issue by assembling a source-independent generalized model that captures syntactic features through part-of-speech and dependency embeddings, as well as contextual features through a fine-tuned language model. We resolve the latter issue by annotating a Twitter dataset which aims at providing a testing ground on a large unstructured dataset. Experimental results show that LESA improves upon the state-of-the-art performance across six benchmark claim datasets by an average of 3 claim-F1 points for in-domain experiments and by 2 claim-F1 points for general-domain experiments. On our dataset too, LESA outperforms existing baselines by 1 claim-F1 point on the in-domain experiments and 2 claim-F1 points on the general-domain experiments. We also release comprehensive data annotation guidelines compiled during the annotation phase (which was missing in the current literature).
Neural models for morphological inflection have recently attained very high results. However, their interpretation remains challenging. Towards this goal, we propose a simple linguistically-motivated variant to the encoder-decoder model with attention. In our model, character-level cross-attention mechanism is complemented with a self-attention module over substrings of the input. We design a novel approach for pattern extraction from attention weights to interpret what the model learn. We apply our methodology to analyze the model’s decisions on three typologically-different languages and find that a) our pattern extraction method applied to cross-attention weights uncovers variation in form of inflection morphemes, b) pattern extraction from self-attention shows triggers for such variation, c) both types of patterns are closely aligned with grammar inflection classes and class assignment criteria, for all three languages. Additionally, we find that the proposed encoder attention component leads to consistent performance improvements over a strong baseline.
Sequence-to-sequence based models have recently shown promising results in generating high-quality questions. However, these models are also known to have main drawbacks such as lack of diversity and bad sentence structures. In this paper, we focus on question generation over SQL database and propose a novel framework by expanding, retrieving, and infilling that first incorporates flexible templates with a neural-based model to generate diverse expressions of questions with sentence structure guidance. Furthermore, a new activation/deactivation mechanism is proposed for template-based sequence-to-sequence generation, which learns to discriminate template patterns and content patterns, thus further improves generation quality. We conduct experiments on two large-scale cross-domain datasets. The experiments show that the superiority of our question generation method in producing more diverse questions while maintaining high quality and consistency under both automatic evaluation and human evaluation.
Word embedding is considered an essential factor in improving the performance of various Natural Language Processing (NLP) models. However, it is hardly applicable in real-world datasets as word embedding is generally studied with a well-refined corpus. Notably, in Hangeul (Korean writing system), which has a unique writing system, various kinds of Out-Of-Vocabulary (OOV) appear from typos. In this paper, we propose a robust Hangeul word embedding model against typos, while maintaining high performance. The proposed model utilizes a Convolutional Neural Network (CNN) architecture with a channel attention mechanism that learns to infer the original word embeddings. The model train with a dataset that consists of a mix of typos and correct words. To demonstrate the effectiveness of the proposed model, we conduct three kinds of intrinsic and extrinsic tasks. While the existing embedding models fail to maintain stable performance as the noise level increases, the proposed model shows stable performance.
This paper addresses the problem of simultaneous machine translation (SiMT) by exploring two main concepts: (a) adaptive policies to learn a good trade-off between high translation quality and low latency; and (b) visual information to support this process by providing additional (visual) contextual information which may be available before the textual input is produced. For that, we propose a multimodal approach to simultaneous machine translation using reinforcement learning, with strategies to integrate visual and textual information in both the agent and the environment. We provide an exploration on how different types of visual information and integration strategies affect the quality and latency of simultaneous translation models, and demonstrate that visual cues lead to higher quality while keeping the latency low.
Mathematical statements written in natural language are usually composed of two different modalities: mathematical elements and natural language. These two modalities have several distinct linguistic and semantic properties. State-of-the-art representation techniques have demonstrated an inability in capturing such an entangled style of discourse. In this work, we propose STAR, a model that uses cross-modal attention to learn how to represent mathematical text for the task of Natural Language Premise Selection. This task uses conjectures written in both natural and mathematical language to recommend premises that most likely will be relevant to prove a particular statement. We found that STAR not only outperforms baselines that do not distinguish between natural language and mathematical elements, but it also achieves better performance than state-of-the-art models.
Multi-hop question answering (QA) requires a model to retrieve and integrate information from multiple passages to answer a question. Rapid progress has been made on multi-hop QA systems with regard to standard evaluation metrics, including EM and F1. However, by simply evaluating the correctness of the answers, it is unclear to what extent these systems have learned the ability to perform multi-hop reasoning. In this paper, we propose an additional sub-question evaluation for the multi-hop QA dataset HotpotQA, in order to shed some light on explaining the reasoning process of QA systems in answering complex questions. We adopt a neural decomposition model to generate sub-questions for a multi-hop question, followed by extracting the corresponding sub-answers. Contrary to our expectation, multiple state-of-the-art multi-hop QA models fail to answer a large portion of sub-questions, although the corresponding multi-hop questions are correctly answered. Our work takes a step forward towards building a more explainable multi-hop QA system.
Recently, it has been found that monolingual English language models can be used as knowledge bases. Instead of structural knowledge base queries, masked sentences such as “Paris is the capital of [MASK]” are used as probes. We translate the established benchmarks TREx and GoogleRE into 53 languages. Working with mBERT, we investigate three questions. (i) Can mBERT be used as a multilingual knowledge base? Most prior work only considers English. Extending research to multiple languages is important for diversity and accessibility. (ii) Is mBERT’s performance as knowledge base language-independent or does it vary from language to language? (iii) A multilingual model is trained on more text, e.g., mBERT is trained on 104 Wikipedias. Can mBERT leverage this for better performance? We find that using mBERT as a knowledge base yields varying performance across languages and pooling predictions across languages improves performance. Conversely, mBERT exhibits a language bias; e.g., when queried in Italian, it tends to predict Italy as the country of origin.
Sentiment analysis is an important task in natural language processing (NLP). Most of existing state-of-the-art methods are under the supervised learning paradigm. However, human annotations can be scarce. Thus, we should leverage more weak supervision for sentiment analysis. In this paper, we propose a posterior regularization framework for the variational approach to the weakly supervised sentiment analysis to better control the posterior distribution of the label assignment. The intuition behind the posterior regularization is that if extracted opinion words from two documents are semantically similar, the posterior distributions of two documents should be similar. Our experimental results show that the posterior regularization can improve the original variational approach to the weakly supervised sentiment analysis and the performance is more stable with smaller prediction variance.
Recent studies treat Word Sense Disambiguation (WSD) as a single-label classification problem in which one is asked to choose only the best-fitting sense for a target word, given its context. However, gold data labelled by expert annotators suggest that maximizing the probability of a single sense may not be the most suitable training objective for WSD, especially if the sense inventory of choice is fine-grained. In this paper, we approach WSD as a multi-label classification problem in which multiple senses can be assigned to each target word. Not only does our simple method bear a closer resemblance to how human annotators disambiguate text, but it can also be seamlessly extended to exploit structured knowledge from semantic networks to achieve state-of-the-art results in English all-words WSD.
Nowadays, fake news is spreading in various ways, and this fake information is causing a lot of social damages. Thus the need to detect fake information is increasing to prevent the damages caused by fake news. In this paper, we propose a novel graph-based fake news detection method using a summarization technique that uses only the document internal information. Our proposed method represents the relationship between all sentences using a graph and the reflection rate of contextual information among sentences is computed by using an attention mechanism. In addition, we improve the performance of fake news detection by utilizing summary information as an important subject of the document.The experimental results demonstrate that our method achieves high accuracy, 91.04%, that is 8.85%p better than the previous method.
Automatic detection of cognates helps downstream NLP tasks of Machine Translation, Cross-lingual Information Retrieval, Computational Phylogenetics and Cross-lingual Named Entity Recognition. Previous approaches for the task of cognate detection use orthographic, phonetic and semantic similarity based features sets. In this paper, we propose a novel method for enriching the feature sets, with cognitive features extracted from human readers’ gaze behaviour. We collect gaze behaviour data for a small sample of cognates and show that extracted cognitive features help the task of cognate detection. However, gaze data collection and annotation is a costly task. We use the collected gaze behaviour data to predict cognitive features for a larger sample and show that predicted cognitive features, also, significantly improve the task performance. We report improvements of 10% with the collected gaze features, and 12% using the predicted gaze features, over the previously proposed approaches. Furthermore, we release the collected gaze behaviour data along with our code and cross-lingual models.
There is a huge difference between a scientific journal reporting ‘wine consumption might be correlated to cancer’, and a media outlet publishing ‘wine causes cancer’ citing the journal’s results. The above example is a typical case of a scientific statement being exaggerated as an outcome of the rising problem of media manipulation. Given a pair of statements (say one from the source journal article and the other from the news article covering the results published in the journal), is it possible to ascertain with some confidence whether one is an exaggerated version of the other? This paper presents a surprisingly simple yet rational three-step approach that performs best for this task. We solve the task by breaking it into three sub-tasks as follows – (a) given a statement from a scientific paper or press release, we first extract relation phrases (e.g., ‘causes’ versus ‘might be correlated to’) connecting the dependent (e.g., ‘cancer’) and the independent (‘wine’) variable, (b) classify the strength of the relationship phrase extracted and (c) compare the strengths of the relation phrases extracted from the statements to identify whether one statement contains an exaggerated version of the other, and to what extent. Through rigorous experiments, we demonstrate that our simple approach by far outperforms baseline models that compare state-of-the-art embedding of the statement pairs through a binary classifier or recast the problem as a textual entailment task, which appears to be a very natural choice in this settings.
Visual dialog is a vision-language task where an agent needs to answer a series of questions grounded in an image based on the understanding of the dialog history and the image. The occurrences of coreference relations in the dialog makes it a more challenging task than visual question-answering. Most previous works have focused on learning better multi-modal representations or on exploring different ways of fusing visual and language features, while the coreferences in the dialog are mainly ignored. In this paper, based on linguistic knowledge and discourse features of human dialog we propose two soft constraints that can improve the model’s ability of resolving coreferences in dialog in an unsupervised way. Experimental results on the VisDial v1.0 dataset shows that our model, which integrates two novel and linguistically inspired soft constraints in a deep transformer neural architecture, obtains new state-of-the-art performance in terms of recall at 1 and other evaluation metrics compared to current existing models and this without pretraining on other vision language datasets. Our qualitative results also demonstrate the effectiveness of the method that we propose.
Neural NLP models tend to rely on spurious correlations between labels and input features to perform their tasks. Minority examples, i.e., examples that contradict the spurious correlations present in the majority of data points, have been shown to increase the out-of-distribution generalization of pre-trained language models. In this paper, we first propose using example forgetting to find minority examples without prior knowledge of the spurious correlations present in the dataset. Forgettable examples are instances either learned and then forgotten during training or never learned. We show empirically how these examples are related to minorities in our training sets. Then, we introduce a new approach to robustify models by fine-tuning our models twice, first on the full training data and second on the minorities only. We obtain substantial improvements in out-of-distribution generalization when applying our approach to the MNLI, QQP and FEVER datasets.
Semantic parsing maps natural language (NL) utterances into logical forms (LFs), which underpins many advanced NLP problems. Semantic parsers gain performance boosts with deep neural networks, but inherit vulnerabilities against adversarial examples. In this paper, we provide the first empirical study on the robustness of semantic parsers in the presence of adversarial attacks. Formally, adversaries of semantic parsing are considered to be the perturbed utterance-LF pairs, whose utterances have exactly the same meanings as the original ones. A scalable methodology is proposed to construct robustness test sets based on existing benchmark corpora. Our results answered five research questions in measuring the sate-of-the-art parsers’ performance on robustness test sets, and evaluating the effect of data augmentation.
Transformer-based “behemoths” have grown in popularity, as well as structurally, shattering multiple NLP benchmarks along the way. However, their real-world usability remains a question. In this work, we empirically assess the feasibility of applying transformer-based models in real-world ad-hoc retrieval applications by comparison to a “greener and more sustainable” alternative, comprising only 620 trainable parameters. We present an analysis of their efficacy and efficiency and show that considering limited computational resources, the lighter model running on the CPU achieves a 3 to 20 times speedup in training and 7 to 47 times in inference while maintaining a comparable retrieval performance. Code to reproduce the efficiency experiments is available on “https://github.com/bioinformatics-ua/EACL2021-reproducibility/“.
When working with problems in natural language processing, we can find ourselves in situations where the traditional measurements of descriptive complexity are ineffective at describing the behaviour of our algorithms. It is easy to see why — the models we use are often general frameworks into which difficult-to-define tasks can be embedded. These frameworks can have more power than we typically use, and so complexity measures such as worst-case running time can drastically overestimate the cost of running our algorithms. In particular, they can make an apparently tractable problem seem NP-complete. Using empirical studies to evaluate performance is a necessary but incomplete method of dealing with this mismatch, since these studies no longer act as a guarantee of good performance. In this paper we use statistical measures such as entropy to give an updated analysis of the complexity of the NP-complete Most Probable Sentence problem for pCFGs, which can then be applied to word sense disambiguation and inference tasks. We can bound both the running time and the error in a simple search algorithm, allowing for a much faster search than the NP-completeness of this problem would suggest.
Although neural models have achieved impressive results on several NLP benchmarks, little is understood about the mechanisms they use to perform language tasks. Thus, much recent attention has been devoted to analyzing the sentence representations learned by neural encoders, through the lens of ‘probing’ tasks. However, to what extent was the information encoded in sentence representations, as discovered through a probe, actually used by the model to perform its task? In this work, we examine this probing paradigm through a case study in Natural Language Inference, showing that models can learn to encode linguistic properties even if they are not needed for the task on which the model was trained. We further identify that pretrained word embeddings play a considerable role in encoding these properties rather than the training task itself, highlighting the importance of careful controls when designing probing experiments. Finally, through a set of controlled synthetic tasks, we demonstrate models can encode these properties considerably above chance-level, even when distributed in the data as random noise, calling into question the interpretation of absolute claims on probing tasks.
This work presents multi-modal deep SVDD (mSVDD) for one-class text classification. By extending the uni-modal SVDD to a multiple modal one, we build mSVDD with multiple hyperspheres, that enable us to build a much better description for target one-class data. Additionally, the end-to-end architecture of mSVDD can jointly handle neural feature learning and one-class text learning. We also introduce a mechanism for incorporating negative supervision in the absence of real negative data, which can be beneficial to the mSVDD model. We conduct experiments on Reuters and 20 Newsgroup datasets, and the experimental results demonstrate that mSVDD outperforms uni-modal SVDD and mSVDD can get further improvements when negative supervision is incorporated.
The number of senses of a given word, or polysemy, is a very subjective notion, which varies widely across annotators and resources. We propose a novel method to estimate polysemy based on simple geometry in the contextual embedding space. Our approach is fully unsupervised and purely data-driven. Through rigorous experiments, we show that our rankings are well correlated, with strong statistical significance, with 6 different rankings derived from famous human-constructed resources such as WordNet, OntoNotes, Oxford, Wikipedia, etc., for 6 different standard metrics. We also visualize and analyze the correlation between the human rankings and make interesting observations. A valuable by-product of our method is the ability to sample, at no extra cost, sentences containing different senses of a given word. Finally, the fully unsupervised nature of our approach makes it applicable to any language. Code and data are publicly available https://github.com/ksipos/polysemy-assessment .
We describe Mega-COV, a billion-scale dataset from Twitter for studying COVID-19. The dataset is diverse (covers 268 countries), longitudinal (goes as back as 2007), multilingual (comes in 100+ languages), and has a significant number of location-tagged tweets (~169M tweets). We release tweet IDs from the dataset. We also develop two powerful models, one for identifying whether or not a tweet is related to the pandemic (best F1=97%) and another for detecting misinformation about COVID-19 (best F1=92%). A human annotation study reveals the utility of our models on a subset of Mega-COV. Our data and models can be useful for studying a wide host of phenomena related to the pandemic. Mega-COV and our models are publicly available.
Spoken language is different from the written language in its style and structure. Disfluencies that appear in transcriptions from speech recognition systems generally hamper the performance of downstream NLP tasks. Thus, a disfluency correction system that converts disfluent to fluent text is of great value. This paper introduces a disfluency correction model that translates disfluent to fluent text by drawing inspiration from recent encoder-decoder unsupervised style-transfer models for text. We also show considerable benefits in performance when utilizing a small sample of 500 parallel disfluent-fluent sentences in a semi-supervised way. Our unsupervised approach achieves a BLEU score of 79.39 on the Switchboard corpus test set, with further improvement to a BLEU score of 85.28 with semi-supervision. Both are comparable to two competitive fully-supervised models.
Question answering (QA) over a knowledge graph (KG) is a task of answering a natural language (NL) query using the information stored in KG. In a real-world industrial setting, this involves addressing multiple challenges including entity linking, multi-hop reasoning over KG, etc. Traditional approaches handle these challenges in a modularized sequential manner where errors in one module lead to the accumulation of errors in downstream modules. Often these challenges are inter-related and the solutions to them can reinforce each other when handled simultaneously in an end-to-end learning setup. To this end, we propose a multi-task BERT based Neural Machine Translation (NMT) model to address these challenges. Through experimental analysis, we demonstrate the efficacy of our proposed approach on one publicly available and one proprietary dataset.
There has been recent success in pre-training on monolingual data and fine-tuning on Machine Translation (MT), but it remains unclear how to best leverage a pre-trained model for a given MT task. This paper investigates the benefits and drawbacks of freezing parameters, and adding new ones, when fine-tuning a pre-trained model on MT. We focus on 1) Fine-tuning a model trained only on English monolingual data, BART. 2) Fine-tuning a model trained on monolingual data from 25 languages, mBART. For BART we get the best performance by freezing most of the model parameters, and adding extra positional embeddings. For mBART we match or outperform the performance of naive fine-tuning for most language pairs with the encoder, and most of the decoder, frozen. The encoder-decoder attention parameters are most important to fine-tune. When constraining ourselves to an out-of-domain training set for Vietnamese to English we see the largest improvements over the fine-tuning baseline.
The distributions of orthographic word types are very different across languages due to typological characteristics, different writing traditions and potentially other factors. The wide range of cross-linguistic diversity is still a major challenge for NLP and the study of language. We use BPE and information-theoretic measures to investigate if distributions become similar under specific levels of subword tokenization. We perform a cross-linguistic comparison, following incremental merges of BPE (we go from characters to words) for 47 diverse languages. We show that text entropy values (a feature of probability distributions) tend to converge at specific subword levels: relatively few BPE merges (around 350) lead to the most similar distributions across languages. Additionally, we analyze the interaction between subword and word-level distributions and show that our findings can be interpreted in light of the ongoing discussion regarding different types of morphological complexity.
We take up the task of large-scale evaluation of neural machine transliteration between English and Indic languages, with a focus on multilingual transliteration to utilize orthographic similarity between Indian languages. We create a corpus of 600K word pairs mined from parallel translation corpora and monolingual corpora, which is the largest transliteration corpora for Indian languages mined from public sources. We perform a detailed analysis of multilingual transliteration and propose an improved multilingual training recipe for Indic languages. We analyze various factors affecting transliteration quality like language family, transliteration direction and word origin.
Formulaic expressions (FEs), such as ‘in this paper, we propose’ are frequently used in scientific papers. FEs convey a communicative function (CF), i.e. ‘showing the aim of the paper’ in the above-mentioned example. Although CF-labelled FEs are helpful in assisting academic writing, the construction of FE databases requires manual labour for assigning CF labels. In this study, we considered a fully automated construction of a CF-labelled FE database using the top–down approach, in which the CF labels are first assigned to sentences, and then the FEs are extracted. For the CF-label assignment, we created a CF-labelled sentence dataset, on which we trained a SciBERT classifier. We show that the classifier and dataset can be used to construct FE databases of disciplines that are different from the training data. The accuracy of in-disciplinary classification was more than 80%, while cross-disciplinary classification also worked well. We also propose an FE extraction method, which was applied to the CF-labelled sentences. Finally, we constructed and published a new, large CF-labelled FE database. The evaluation of the final CF-labelled FE database showed that approximately 65% of the FEs are correct and useful, which is sufficiently high considering practical use.
Major scandals in corporate history have urged the need for regulatory compliance, where organizations need to ensure that their controls (processes) comply with relevant laws, regulations, and policies. However, keeping track of the constantly changing legislation is difficult, thus organizations are increasingly adopting Regulatory Technology (RegTech) to facilitate the process. To this end, we introduce regulatory information retrieval (REG-IR), an application of document-to-document information retrieval (DOC2DOC IR), where the query is an entire document making the task more challenging than traditional IR where the queries are short. Furthermore, we compile and release two datasets based on the relationships between EU directives and UK legislation. We experiment on these datasets using a typical two-step pipeline approach comprising a pre-fetcher and a neural re-ranker. Experimenting with various pre-fetchers from BM25 to k nearest neighbors over representations from several BERT models, we show that fine-tuning a BERT model on an in-domain classification task produces the best representations for IR. We also show that neural re-rankers under-perform due to contradicting supervision, i.e., similar query-document pairs with opposite labels. Thus, they are biased towards the pre-fetcher’s score. Interestingly, applying a date filter further improves the performance, showcasing the importance of the time dimension.
The growing availability of powerful mobile devices and other edge devices, together with increasing regulatory and security concerns about the exchange of personal information across networks of these devices has challenged the Computational Linguistics community to develop methods that are at once fast, space-efficient, accurate and amenable to secure encoding schemes such as homomorphic encryption. Inspired by recent work that restricts floating point precision to speed up neural network training in hardware-based SIMD, we have developed a method for compressing word vector embeddings into integers using the Chinese Reminder Theorem that speeds up addition by up to 48.27% and at the same time compresses GloVe word embedding libraries by up to 25.86%. We explore the practicality of this simple approach by investigating the trade-off between precision and performance in two NLP tasks: compositional semantic relatedness and opinion target sentiment classification. We find that in both tasks, lowering floating point number precision results in negligible changes to performance.
In the paraphrase generation task, source sentences often contain phrases that should not be altered. Which phrases, however, can be context dependent and can vary by application. Our solution to this challenge is to provide the user with explicit tags that can be placed around any arbitrary segment of text to mean “don’t change me!” when generating a paraphrase; the model learns to explicitly copy these phrases to the output. The contribution of this work is a novel data generation technique using distant supervision that allows us to start with a pretrained sequence-to-sequence model and fine-tune a paraphrase generator that exhibits this behavior, allowing user-controllable paraphrase generation. Additionally, we modify the loss during fine-tuning to explicitly encourage diversity in model output. Our technique is language agnostic, and we report experiments in English and Chinese.
Although coherence modeling has come a long way in developing novel models, their evaluation on downstream applications for which they are purportedly developed has largely been neglected. With the advancements made by neural approaches in applications such as machine translation (MT), summarization and dialog systems, the need for coherence evaluation of these tasks is now more crucial than ever. However, coherence models are typically evaluated only on synthetic tasks, which may not be representative of their performance in downstream applications. To investigate how representative the synthetic tasks are of downstream use cases, we conduct experiments on benchmarking well-known traditional and neural coherence models on synthetic sentence ordering tasks, and contrast this with their performance on three downstream applications: coherence evaluation for MT and summarization, and next utterance prediction in retrieval-based dialog. Our results demonstrate a weak correlation between the model performances in the synthetic tasks and the downstream applications, motivating alternate training and evaluation methods for coherence models.
Political discussions revolve around ideological conflicts that often split the audience into two opposing parties. Both parties try to win the argument by bringing forward information. However, often this information is misleading, and its dissemination employs propaganda techniques. In this work, we analyze the impact of propaganda on six major political forums on Reddit that target a diverse audience in two countries, the US and the UK. We focus on three research questions: who is posting propaganda? how does propaganda differ across the political spectrum? and how is propaganda received on political forums?
Contextualised word representation models have been successfully used for capturing different word usages and they may be an attractive alternative for representing idiomaticity in language. In this paper, we propose probing measures to assess if some of the expected linguistic properties of noun compounds, especially those related to idiomatic meanings, and their dependence on context and sensitivity to lexical choice, are readily available in some standard and widely used representations. For that, we constructed the Noun Compound Senses Dataset, which contains noun compounds and their paraphrases, in context neutral and context informative naturalistic sentences, in two languages: English and Portuguese. Results obtained using four types of probing measures with models like ELMo, BERT and some of its variants, indicate that idiomaticity is not yet accurately represented by contextualised models
An in-depth analysis of the level of language understanding required by existing Machine Reading Comprehension (MRC) benchmarks can provide insight into the reading capabilities of machines. In this paper, we propose an ablation-based methodology to assess the extent to which MRC datasets evaluate the understanding of explicit discourse relations. We define seven MRC skills which require the understanding of different discourse relations. We then introduce ablation methods that verify whether these skills are required to succeed on a dataset. By observing the drop in performance of neural MRC models evaluated on the original and the modified dataset, we can measure to what degree the dataset requires these skills, in order to be understood correctly. Experiments on three large-scale datasets with the BERT-base and ALBERT-xxlarge model show that the relative changes for all skills are small (less than 6%). These results imply that most of the answered questions in the examined datasets do not require understanding the discourse structure of the text. To specifically probe for natural language understanding, there is a need to design more challenging benchmarks that can correctly evaluate the intended skills.
Automatic detection of the four MBTI personality dimensions from texts has recently attracted noticeable attention from the natural language processing and computational linguistic communities. Despite the large collections of Twitter data for training, the best systems rarely even outperform the majority-class baseline. In this paper, we discuss the theoretical reasons for such low results and present the insights from an annotation study that further shed the light on this issue.
Pretrained using large amount of data, autoregressive language models are able to generate high quality sequences. However, these models do not perform well under hard lexical constraints as they lack fine control of content generation process. Progressive insertion based transformers can overcome the above limitation and efficiently generate a sequence in parallel given some input tokens as constraint. These transformers however may fail to support hard lexical constraints as their generation process is more likely to terminate prematurely. The paper analyses such early termination problems and proposes the ENtity CONstrained insertion TransformER(ENCONTER), a new insertion transformer that addresses the above pitfall without compromising much generation efficiency. We introduce a new training strategy that considers predefined hard lexical constraints (e.g., entities to be included in the generated sequence). Our experiments show that ENCONTER outperforms other baseline models in several performance metrics rendering it more suitable in practical applications.
Natural language processing (NLP) tasks (e.g. question-answering in English) benefit from knowledge of other tasks (e.g., named entity recognition in English) and knowledge of other languages (e.g., question-answering in Spanish). Such shared representations are typically learned in isolation, either across tasks or across languages. In this work, we propose a meta-learning approach to learn the interactions between both tasks and languages. We also investigate the role of different sampling strategies used during meta-learning. We present experiments on five different tasks and six different languages from the XTREME multilingual benchmark dataset. Our meta-learned model clearly improves in performance compared to competitive baseline models that also include multi-task baselines. We also present zero-shot evaluations on unseen target languages to demonstrate the utility of our proposed model.
Several machine learning-based spoiler detection models have been proposed recently to protect users from spoilers on review websites. Although dependency relations between context words are important for detecting spoilers, current attention-based spoiler detection models are insufficient for utilizing dependency relations. To address this problem, we propose a new spoiler detection model called SDGNN that is based on syntax-aware graph neural networks. In the experiments on two real-world benchmark datasets, we show that our SDGNN outperforms the existing spoiler detection models.
Large pre-trained language models have been shown to encode large amounts of world and commonsense knowledge in their parameters, leading to substantial interest in methods for extracting that knowledge. In past work, knowledge was extracted by taking manually-authored queries and gathering paraphrases for them using a separate pipeline. In this work, we propose a method for automatically rewriting queries into “BERTese”, a paraphrase query that is directly optimized towards better knowledge extraction. To encourage meaningful rewrites, we add auxiliary loss functions that encourage the query to correspond to actual language tokens. We empirically show our approach outperforms competing baselines, obviating the need for complex pipelines. Moreover, BERTese provides some insight into the type of language that helps language models perform knowledge extraction.
The ability of humans to symbolically represent social events and situations is crucial for various interactions in everyday life. Several studies in cognitive psychology have established the role of mental state attributions in effectively representing variable aspects of these social events. In the past, NLP research on learning event representations often focuses on construing syntactic and semantic information from language. However, they fail to consider the importance of pragmatic aspects and the need to consistently update new social situational information without forgetting the accumulated experiences. In this work, we propose a representation learning framework to directly address these shortcomings by integrating social commonsense knowledge with recent advancements in the space of lifelong language learning. First, we investigate methods to incorporate pragmatic aspects into our social event embeddings by leveraging social commonsense knowledge. Next, we introduce continual learning strategies that allow for incremental consolidation of new knowledge while retaining and promoting efficient usage of prior knowledge. Experimental results on event similarity, reasoning, and paraphrase detection tasks prove the efficacy of our social event embeddings.
Instead of using expensive manual annotations, researchers have proposed to train named entity recognition (NER) systems using heuristic labeling rules. However, devising labeling rules is challenging because it often requires a considerable amount of manual effort and domain expertise. To alleviate this problem, we propose GLARA, a graph-based labeling rule augmentation framework, to learn new labeling rules from unlabeled data. We first create a graph with nodes representing candidate rules extracted from unlabeled data. Then, we design a new graph neural network to augment labeling rules by exploring the semantic relations between rules. We finally apply the augmented rules on unlabeled data to generate weak labels and train a NER model using the weakly labeled data. We evaluate our method on three NER datasets and find that we can achieve an average improvement of +20% F1 score over the best baseline when given a small set of seed rules.
We present a joint model for entity-level relation extraction from documents. In contrast to other approaches - which focus on local intra-sentence mention pairs and thus require annotations on mention level - our model operates on entity level. To do so, a multi-task approach is followed that builds upon coreference resolution and gathers relevant signals via multi-instance learning with multi-level representations combining global entity and local mention information. We achieve state-of-the-art relation extraction results on the DocRED dataset and report the first entity-level end-to-end relation extraction results for future reference. Finally, our experimental results suggest that a joint approach is on par with task-specific learning, though more efficient due to shared parameters and training steps.
Automatic Speech Recognition (ASR) systems are evaluated using Word Error Rate (WER), which is calculated by comparing the number of errors between the ground truth and the transcription of the ASR system. This calculation, however, requires manual transcription of the speech signal to obtain the ground truth. Since transcribing audio signals is a costly process, Automatic WER Evaluation (e-WER) methods have been developed to automatically predict the WER of a speech system by only relying on the transcription and the speech signal features. While WER is a continuous variable, previous works have shown that positing e-WER as a classification problem is more effective than regression. However, while converting to a classification setting, these approaches suffer from heavy class imbalance. In this paper, we propose a new balanced paradigm for e-WER in a classification setting. Within this paradigm, we also propose WER-BERT, a BERT based architecture with speech features for e-WER. Furthermore, we introduce a distance loss function to tackle the ordinal nature of e-WER classification. The proposed approach and paradigm are evaluated on the Librispeech dataset and a commercial (black box) ASR system, Google Cloud’s Speech-to-Text API. The results and experiments demonstrate that WER-BERT establishes a new state-of-the-art in automatic WER estimation.
This paper explores how the Distantly Supervised Relation Extraction (DS-RE) can benefit from the use of a Universal Graph (UG), the combination of a Knowledge Graph (KG) and a large-scale text collection. A straightforward extension of a current state-of-the-art neural model for DS-RE with a UG may lead to degradation in performance. We first report that this degradation is associated with the difficulty in learning a UG and then propose two training strategies: (1) Path Type Adaptive Pretraining, which sequentially trains the model with different types of UG paths so as to prevent the reliance on a single type of UG path; and (2) Complexity Ranking Guided Attention mechanism, which restricts the attention span according to the complexity of a UG path so as to force the model to extract features not only from simple UG paths but also from complex ones. Experimental results on both biomedical and NYT10 datasets prove the robustness of our methods and achieve a new state-of-the-art result on the NYT10 dataset. The code and datasets used in this paper are available at https://github.com/baodaiqin/UGDSRE.
Automatic Post-Editing (APE) aims to correct errors in the output of a given machine translation (MT) system. Although data-driven approaches have become prevalent also in the APE task as in many other NLP tasks, there has been a lack of qualified training data due to the high cost of manual construction. eSCAPE, a synthetic APE corpus, has been widely used to alleviate the data scarcity, but it might not address genuine APE corpora’s characteristic that the post-edited sentence should be a minimally edited revision of the given MT output. Therefore, we propose two new methods of synthesizing additional MT outputs by adapting back-translation to the APE task, obtaining robust enlargements of the existing synthetic APE training dataset. Experimental results on the WMT English-German APE benchmarks demonstrate that our enlarged datasets are effective in improving APE performance.
Unsupervised image captioning is a challenging task that aims at generating captions without the supervision of image-sentence pairs, but only with images and sentences drawn from different sources and object labels detected from the images. In previous work, pseudo-captions, i.e., sentences that contain the detected object labels, were assigned to a given image. The focus of the previous work was on the alignment of input images and pseudo-captions at the sentence level. However, pseudo-captions contain many words that are irrelevant to a given image. In this work, we investigate the effect of removing mismatched words from image-sentence alignment to determine how they make this task difficult. We propose a simple gating mechanism that is trained to align image features with only the most reliable words in pseudo-captions: the detected object labels. The experimental results show that our proposed method outperforms the previous methods without introducing complex sentence-level learning objectives. Combined with the sentence-level alignment method of previous work, our method further improves its performance. These results confirm the importance of careful alignment in word-level details.
Performance prediction, the task of estimating a system’s performance without performing experiments, allows us to reduce the experimental burden caused by the combinatorial explosion of different datasets, languages, tasks, and models. In this paper, we make two contributions to improving performance prediction for NLP tasks. First, we examine performance predictors not only for holistic measures of accuracy like F1 or BLEU, but also fine-grained performance measures such as accuracy over individual classes of examples. Second, we propose methods to understand the reliability of a performance prediction model from two angles: confidence intervals and calibration. We perform an analysis of four types of NLP tasks, and both demonstrate the feasibility of fine-grained performance prediction and the necessity to perform reliability analysis for performance prediction methods in the future.
A prerequisite for the computational study of literature is the availability of properly digitized texts, ideally with reliable meta-data and ground-truth annotation. Poetry corpora do exist for a number of languages, but larger collections lack consistency and are encoded in various standards, while annotated corpora are typically constrained to a particular genre and/or were designed for the analysis of certain linguistic features (like rhyme). In this work, we provide large poetry corpora for English and German, and annotate prosodic features in smaller corpora to train corpus driven neural models that enable robust large scale analysis. We show that BiLSTM-CRF models with syllable embeddings outperform a CRF baseline and different BERT-based approaches. In a multi-task setup, particular beneficial task relations illustrate the inter-dependence of poetic features. A model learns foot boundaries better when jointly predicting syllable stress, aesthetic emotions and verse measures benefit from each other, and we find that caesuras are quite dependent on syntax and also integral to shaping the overall measure of the line.
Aspect-level sentiment analysis (ASA) has received much attention in recent years. Most existing approaches tried to leverage syntactic information, such as the dependency parsing results of the input text, to improve sentiment analysis on different aspects. Although these approaches achieved satisfying results, their main focus is to leverage the dependency arcs among words where the dependency type information is omitted; and they model different dependencies equally where the noisy dependency results may hurt model performance. In this paper, we propose an approach to enhance aspect-level sentiment analysis with word dependencies, where the type information is modeled by key-value memory networks and different dependency results are selectively leveraged. Experimental results on five benchmark datasets demonstrate the effectiveness of our approach, where it outperforms baseline models on all datasets and achieves state-of-the-art performance on three of them.
I present rst-workbench, a software package that simplifies the installation and usage of numerous end-to-end Rhetorical Structure Theory (RST) parsers. The tool offers a web-based interface that allows users to enter text and let multiple RST parsers generate analyses concurrently. The resulting RST trees can be compared visually, manually post-edited (in the browser) and stored for later usage.
Although open-domain question answering (QA) draws great attention in recent years, it requires large amounts of resources for building the full system and it is often difficult to reproduce previous results due to complex configurations. In this paper, we introduce SF-QA: simple and fair evaluation framework for open-domain QA. SF-QA framework modularizes the pipeline open-domain QA system, which makes the task itself easily accessible and reproducible to research groups without enough computing resources. The proposed evaluation framework is publicly available and anyone can contribute to the code and evaluations.
This paper presents an open-source library for efficient low-level processing of ten major South Asian Brahmic scripts. The library provides a flexible and extensible framework for supporting crucial operations on Brahmic scripts, such as NFC, visual normalization, reversible transliteration, and validity checks, implemented in Python within a finite-state transducer formalism. We survey some common Brahmic script issues that may adversely affect the performance of downstream NLP tasks, and provide the rationale for finite-state design and system implementation details.
This paper presents CovRelex, a scientific paper retrieval system targeting entities and relations via relation extraction on COVID-19 scientific papers. This work aims at building a system supporting users efficiently in acquiring knowledge across a huge number of COVID-19 scientific papers published rapidly. Our system can be accessed via https://www.jaist.ac.jp/is/labs/nguyen-lab/systems/covrelex/.
Dialogue Systems are becoming ubiquitous in various forms and shapes - virtual assistants(Siri, Alexa, etc.), chat-bots, customer sup-port, chit-chat systems just to name a few.The advances in language models and their publication have democratised advanced NLP.However, data remains a crucial bottleneck.Our contribution to this essential pillar isMATILDA, to the best of our knowledge the first multi-annotator, multi-language dialogue annotation tool. MATILDA allows the creation of corpora, the management of users, the annotation of dialogues, the quick adaptation of the user interface to any language and the resolution of inter-annotator disagreement. We evaluate the tool on ease of use, annotation speed and interannotation resolution for both experts and novices and conclude that this tool not only supports the full pipeline for dialogue annotation, but also allows non-technical people to easily use it. We are completely open-sourcing the tool at https://github.com/wluper/matilda and provide a tutorial video1.
One strategy for facilitating reading comprehension is to present information in a question-and-answer format. We demo a system that integrates the tasks of question answering (QA) and question generation (QG) in order to produce Q&A items that convey the content of multi-paragraph documents. We report some experiments for QA and QG that yield improvements on both tasks, and assess how they interact to produce a list of Q&A items for a text. The demo is accessible at qna.sdl.com.
Language model (LM) pretraining has led to consistent improvements in many NLP downstream tasks, including named entity recognition (NER). In this paper, we present T-NER (Transformer-based Named Entity Recognition), a Python library for NER LM finetuning. In addition to its practical utility, T-NER facilitates the study and investigation of the cross-domain and cross-lingual generalization ability of LMs finetuned on NER. Our library also provides a web app where users can get model predictions interactively for arbitrary text, which facilitates qualitative model evaluation for non-expert programmers. We show the potential of the library by compiling nine public NER datasets into a unified format and evaluating the cross-domain and cross- lingual performance across the datasets. The results from our initial experiments show that in-domain performance is generally competitive across datasets. However, cross-domain generalization is challenging even with a large pretrained LM, which has nevertheless capacity to learn domain-specific features if fine- tuned on a combined dataset. To facilitate future research, we also release all our LM checkpoints via the Hugging Face model hub.
With the increasing number of user comments in diverse domains, including comments on online journalism and e-commerce websites, the manual content analysis of these comments becomes time-consuming and challenging. However, research showed that user comments contain useful information for different domain experts, which is thus worth finding and utilizing. This paper introduces Forum 4.0, an open-source framework to semi-automatically analyze, aggregate, and visualize user comments based on labels defined by domain experts. We demonstrate the applicability of Forum 4.0 with comments analytics scenarios within the domains of online journalism and app stores. We outline the underlying container architecture, including the web-based user interface, the machine learning component, and the task manager for time-consuming tasks. We finally conduct machine learning experiments with simulated annotations and different sampling strategies on existing datasets from both domains to evaluate Forum 4.0’s performance. Forum 4.0 achieves promising classification results (ROC-AUC ≥ 0.9 with 100 annotated samples), utilizing transformer-based embeddings with a lightweight logistic regression model. We explain how Forum 4.0’s architecture is applicable for millions of user comments in real-time, yet at feasible training and classification costs.
Automatic evaluation of Machine Translation (MT) quality has been investigated over several decades. Spoken Language Translation (SLT), esp. when simultaneous, needs to consider additional criteria and does not have a standard evaluation procedure and a widely used toolkit. To fill the gap, we develop SLTev, an open-source tool for assessing SLT in a comprehensive way. SLTev reports the quality, latency, and stability of an SLT candidate output based on the time-stamped transcript and reference translation into a target language. For quality, we rely on sacreBLEU which provides MT evaluation measures such as chrF or BLEU. For latency, we propose two new scoring techniques. For stability, we extend the previously defined measures with a normalized Flicker in our work. We also propose a new averaging of older measures. A preliminary version of SLTev was used in the IWSLT 2020 shared task. Moreover, a growing collection of test datasets directly accessible by SLTev are provided for system evaluation comparable across papers.
We introduce Trankit, a light-weight Transformer-based Toolkit for multilingual Natural Language Processing (NLP). It provides a trainable pipeline for fundamental NLP tasks over 100 languages, and 90 pretrained pipelines for 56 languages. Built on a state-of-the-art pretrained language model, Trankit significantly outperforms prior multilingual NLP pipelines over sentence segmentation, part-of-speech tagging, morphological feature tagging, and dependency parsing while maintaining competitive performance for tokenization, multi-word token expansion, and lemmatization over 90 Universal Dependencies treebanks. Despite the use of a large pretrained transformer, our toolkit is still efficient in memory usage and speed. This is achieved by our novel plug-and-play mechanism with Adapters where a multilingual pretrained transformer is shared across pipelines for different languages. Our toolkit along with pretrained models and code are publicly available at: https://github.com/nlp-uoregon/trankit. A demo website for our toolkit is also available at: http://nlp.uoregon.edu/trankit. Finally, we create a demo video for Trankit at: https://youtu.be/q0KGP3zGjGc.
Recent research efforts in NLP have demonstrated that distributional word vector spaces often encode stereotypical human biases, such as racism and sexism. With word representations ubiquitously used in NLP models and pipelines, this raises ethical issues and jeopardizes the fairness of language technologies. While there exists a large body of work on bias measures and debiasing methods, to date, there is no platform that would unify these research efforts and make bias measuring and debiasing of representation spaces widely accessible. In this work, we present DebIE, the first integrated platform for (1) measuring and (2) mitigating bias in word embeddings. Given an (i) embedding space (users can choose between the predefined spaces or upload their own) and (ii) a bias specification (users can choose between existing bias specifications or create their own), DebIE can (1) compute several measures of implicit and explicit bias and modify the embedding space by executing two (mutually composable) debiasing models. DebIE’s functionality can be accessed through four different interfaces: (a) a web application, (b) a desktop application, (c) a REST-ful API, and (d) as a command-line application. DebIE is available at: debie.informatik.uni-mannheim.de.
This paper describes the current milestones achieved in our ongoing project that aims to understand the surveillance of, impact of and intervention on COVID-19 misinfodemic on Twitter. Specifically, it introduces a public dashboard which, in addition to displaying case counts in an interactive map and a navigational panel, also provides some unique features not found in other places. Particularly, the dashboard uses a curated catalog of COVID-19 related facts and debunks of misinformation, and it displays the most prevalent information from the catalog among Twitter users in user-selected U.S. geographic regions. The paper explains how to use BERT models to match tweets with the facts and misinformation and to detect their stance towards such information. The paper also discusses the results of preliminary experiments on analyzing the spatio-temporal spread of misinformation.
Amazon Mechanical Turk (AMT) has recently become one of the most popular crowd-sourcing platforms, allowing researchers from all over the world to create linguistic datasets quickly and at a relatively low cost. Amazon provides both a web interface and an API for AMT, but they are not very user-friendly and miss some features that can be useful for NLP researchers. In this paper, we present EasyTurk, a free tool that improves the potential of Amazon Mechanical Turk by adding to it some new features. The tool is free and released under an open source license.
This system demonstration paper describes ASAD: Arabic Social media Analysis and unDerstanding, a suite of seven individual modules that allows users to determine dialects, sentiment, news category, offensiveness, hate speech, adult content, and spam in Arabic tweets. The suite is made available through a web API and a web interface where users can enter text or upload files.
In this paper we present COCO-EX, a tool for Extracting Concepts from texts and linking them to the ConceptNet knowledge graph. COCO-EX extracts meaningful concepts from natural language texts and maps them to conjunct concept nodes in ConceptNet, utilizing the maximum of relational information stored in the ConceptNet knowledge graph. COCOEX takes into account the challenging characteristics of ConceptNet, namely that – unlike conventional knowledge graphs – nodes are represented as non-canonicalized, free-form text. This means that i) concepts are not normalized; ii) they often consist of several different, nested phrase types; and iii) many of them are uninformative, over-specific, or misspelled. A commonly used shortcut to circumvent these problems is to apply string matching. We compare COCO-EX to this method and show that COCO-EX enables the extraction of meaningful, important rather than overspecific or uninformative concepts, and allows to assess more relational information stored in the knowledge graph.
Several tools and resources have been developed to deal with Arabic NLP. However, a homogenous and flexible Arabic environment that gathers these components is rarely available. In this perspective, we introduce SAFAR which is a monolingual framework developed in accordance with software engineering requirements and dedicated to Arabic language, especially, the modern standard Arabic and Moroccan dialect. After one decade of integration and development, SAFAR possesses today more than 50 tools and resources that can be exploited either using its API or using its web interface.
With the increasingly widespread use of Transformer-based models for NLU/NLP tasks, there is growing interest in understanding the inner workings of these models, why they are so effective at a wide range of tasks, and how they can be further tuned and improved. To contribute towards this goal of enhanced explainability and comprehension, we present InterpreT, an interactive visualization tool for interpreting Transformer-based models. In addition to providing various mechanisms for investigating general model behaviours, novel contributions made in InterpreT include the ability to track and visualize token embeddings through each layer of a Transformer, highlight distances between certain token embeddings through illustrative plots, and identify task-related functions of attention heads by using new metrics. InterpreT is a task agnostic tool, and its functionalities are demonstrated through the analysis of model behaviours for two disparate tasks: Aspect Based Sentiment Analysis (ABSA) and the Winograd Schema Challenge (WSC).
We describe a new addition to the WebVectors toolkit which is used to serve word embedding models over the Web. The new ELMoViz module adds support for contextualized embedding architectures, in particular for ELMo models. The provided visualizations follow the metaphor of ‘two-dimensional text’ by showing lexical substitutes: words which are most semantically similar in context to the words of the input sentence. The system allows the user to change the ELMo layers from which token embeddings are inferred. It also conveys corpus information about the query words and their lexical substitutes (namely their frequency tiers and parts of speech). The module is well integrated into the rest of the WebVectors toolkit, providing lexical hyperlinks to word representations in static embedding models. Two web services have already implemented the new functionality with pre-trained ELMo models for Russian, Norwegian and English.
We present LOME, a system for performing multilingual information extraction. Given a text document as input, our core system identifies spans of textual entity and event mentions with a FrameNet (Baker et al., 1998) parser. It subsequently performs coreference resolution, fine-grained entity typing, and temporal relation prediction between events. By doing so, the system constructs an event and entity focused knowledge graph. We can further apply third-party modules for other types of annotation, like relation extraction. Our (multilingual) first-party modules either outperform or are competitive with the (monolingual) state-of-the-art. We achieve this through the use of multilingual encoders like XLM-R (Conneau et al., 2020) and leveraging multilingual training data. LOME is available as a Docker container on Docker Hub. In addition, a lightweight version of the system is accessible as a web demo.
Acronyms and abbreviations are the short-form of longer phrases and they are ubiquitously employed in various types of writing. Despite their usefulness to save space in writing and reader’s time in reading, they also provide challenges for understanding the text especially if the acronym is not defined in the text or if it is used far from its definition in long texts. To alleviate this issue, there are considerable efforts both from the research community and software developers to build systems for identifying acronyms and finding their correct meanings in the text. However, none of the existing works provide a unified solution capable of processing acronyms in various domains and to be publicly available. Thus, we provide the first web-based acronym identification and disambiguation system which can process acronyms from various domains including scientific, biomedical, and general domains. The web-based system is publicly available at http://iq.cs.uoregon.edu:5000 and a demo video is available at https://youtu.be/IkSh7LqI42M. The system source code is also available at https://github.com/amirveyseh/MadDog.
This article presents a set of tools built around the Graph Rewriting computational framework which can be used to compute complex rule-based transformations on linguistic structures. Application of the graph matching mechanism for corpus exploration, error mining or quantitative typology are also given.
Transfer learning, particularly approaches that combine multi-task learning with pre-trained contextualized embeddings and fine-tuning, have advanced the field of Natural Language Processing tremendously in recent years. In this paper we present MaChAmp, a toolkit for easy fine-tuning of contextualized embeddings in multi-task settings. The benefits of MaChAmp are its flexible configuration options, and the support of a variety of natural language processing tasks in a uniform toolkit, from text classification and sequence labeling to dependency parsing, masked language modeling, and text generation.
We present Sense Clustering over Time (SCoT), a novel network-based tool for analysing lexical change. SCoT represents the meanings of a word as clusters of similar words. It visualises their formation, change, and demise. There are two main approaches to the exploration of dynamic networks: the discrete one compares a series of clustered graphs from separate points in time. The continuous one analyses the changes of one dynamic network over a time-span. SCoT offers a new hybrid solution. First, it aggregates time-stamped documents into intervals and calculates one sense graph per discrete interval. Then, it merges the static graphs to a new type of dynamic semantic neighbourhood graph over time. The resulting sense clusters offer uniquely detailed insights into lexical change over continuous intervals with model transparency and provenance. SCoT has been successfully used in a European study on the changing meaning of ‘crisis’.
Code-mixing is common in multilingual communities around the world, and processing it is challenging due to the lack of labeled and unlabeled data. We describe a tool that can automatically generate code-mixed data given parallel data in two languages. We implement two linguistic theories of code-mixing, the Equivalence Constraint theory and the Matrix Language theory to generate all possible code-mixed sentences in the language-pair, followed by sampling of the generated data to generate natural code-mixed sentences. The toolkit provides three modes: a batch mode, an interactive library mode and a web-interface to address the needs of researchers, linguists and language experts. The toolkit can be used to generate unlabeled text data for pre-trained models, as well as visualize linguistic theories of code-mixing. We plan to release the toolkit as open source and extend it by adding more implementations of linguistic theories, visualization techniques and better sampling techniques. We expect that the release of this toolkit will help facilitate more research in code-mixing in diverse language pairs.
Recent advances in deep transformer models have achieved state-of-the-art in several natural language processing (NLP) tasks, whereas named entity recognition (NER) has traditionally benefited from long-short term memory (LSTM) networks. In this work, we present a Transformers based Transfer Learning framework for Named Entity Recognition (T2NER) created in PyTorch for the task of NER with deep transformer models. The framework is built upon the Transformers library as the core modeling engine and supports several transfer learning scenarios from sequential transfer to domain adaptation, multi-task learning, and semi-supervised learning. It aims to bridge the gap between the algorithmic advances in these areas by combining them with the state-of-the-art in transformer models to provide a unified platform that is readily extensible and can be used for both the transfer learning research in NER, and for real-world applications. The framework is available at: https://github.com/suamin/t2ner.
Europe is a multilingual society, in which dozens of languages are spoken. The only option to enable and to benefit from multilingualism is through Language Technologies (LT), i.e., Natural Language Processing and Speech Technologies. We describe the European Language Grid (ELG), which is targeted to evolve into the primary platform and marketplace for LT in Europe by providing one umbrella platform for the European LT landscape, including research and industry, enabling all stakeholders to upload, share and distribute their services, products and resources. At the end of our EU project, which will establish a legal entity in 2022, the ELG will provide access to approx. 1300 services for all European languages as well as thousands of data sets.
One way to enhance user engagement in search engines is to suggest interesting facts to the user. Although relationships between persons are important as a target for text mining, there are few effective approaches for extracting the interesting relationships between persons. We therefore propose a method for extracting interesting relationships between persons from natural language texts by focusing on their surprisingness. Our method first extracts all personal relationships from dependency trees for the texts and then calculates surprise scores for distributed representations of the extracted relationships in an unsupervised manner. The unique point of our method is that it does not require any labeled dataset with annotation for the surprising personal relationships. The results of the human evaluation show that the proposed method could extract more interesting relationships between persons from Japanese Wikipedia articles than a popularity-based baseline method. We demonstrate our proposed method as a chrome plugin on google search.
In this paper, we present Paladin, an open-source web-based annotation tool for creating high-quality multi-label document-level datasets. By integrating active learning and proactive learning to the annotation task, Paladin makes the task less time-consuming and requiring less human effort. Although Paladin is designed for multi-label settings, the system is flexible and can be adapted to other tasks in single-label settings.
Few shot learning with large language models has the potential to give individuals without formal machine learning training the access to a wide range of text to text models. We consider how this applies to creative writers and present Story Centaur, a user interface for prototyping few shot models and a set of recombinable web components that deploy them. Story Centaur’s goal is to expose creative writers to few shot learning with a simple but powerful interface that lets them compose their own co-creation tools that further their own unique artistic directions. We build out several examples of such tools, and in the process probe the boundaries and issues surrounding generation with large language models.
In this paper, we introduce FrameForm, an open-source annotation tool designed to accommodate predicate annotations based on Frame Semantics. FrameForm is a user-friendly tool for creating, annotating and maintaining computational lexicography projects like FrameNet and has been used while building the Turkish FrameNet. Responsive and open-source, FrameForm can be easily modified to answer the annotation needs of a wide range of different languages.
In this paper, we present OCTIS, a framework for training, analyzing, and comparing Topic Models, whose optimal hyper-parameters are estimated using a Bayesian Optimization approach. The proposed solution integrates several state-of-the-art topic models and evaluation metrics. These metrics can be targeted as objective by the underlying optimization procedure to determine the best hyper-parameter configuration. OCTIS allows researchers and practitioners to have a fair comparison between topic models of interest, using several benchmark datasets and well-known evaluation metrics, to integrate novel algorithms, and to have an interactive visualization of the results for understanding the behavior of each model. The code is available at the following link: https://github.com/MIND-Lab/OCTIS.
This paper presents an automatic speech translation system aimed at live subtitling of conference presentations. We describe the overall architecture and key processing components. More importantly, we explain our strategy for building a complex system for end-users from numerous individual components, each of which has been tested only in laboratory conditions. The system is a working prototype that is routinely tested in recognizing English, Czech, and German speech and presenting it translated simultaneously into 42 target languages.
It is standard procedure these days to solve Information Extraction task by fine-tuning large pre-trained language models. This is not the case for generation task, which relies on a variety of techniques for controlled language generation. In this paper, we describe a system that fine-tunes a natural language generation model for the problem of solving writer’s block. The fine-tuning changes the conditioning to also include the right context in addition to the left context, as well as an optional list of entities, the size, the genre and a summary of the paragraph that the human author wishes to generate. Our proposed fine-tuning obtains excellent results, even with a small number of epochs and a total cost of USD 150. The system can be accessed as a web-service and all the code is released. A video showcasing the interface and the model is also available.
OPUS-CAT is a collection of software which enables translators to use neural machine translation in computer-assisted translation tools without exposing themselves to security and confidentiality risks inherent in online machine translation. OPUS-CAT uses the public OPUS-MT machine translation models, which are available for over a thousand language pairs. The generic OPUS-MT models can be fine-tuned with OPUS-CAT on the desktop using data for a specific client or domain.
Today, most dialogue systems are fully or partly built using neural network architectures. A crucial prerequisite for the creation of a goal-oriented neural network dialogue system is a dataset that represents typical dialogue scenarios and includes various semantic annotations, e.g. intents, slots and dialogue actions, that are necessary for training a particular neural network architecture. In this demonstration paper, we present an easy to use interface and its back-end which is oriented to domain experts for the collection of goal-oriented dialogue samples. The platform not only allows to collect or write sample dialogues in a structured way, but also provides a means for simple annotation and interpretation of the dialogues. The platform itself is language-independent; it depends only on the availability of particular language processing components for a specific language. It is currently being used to collect dialogue samples in Latvian (a highly inflected language) which represent typical communication between students and the student service.
We present a system for answering comparative questions (Is X better than Y with respect to Z?) in natural language. Answering such questions is important for assisting humans in making informed decisions. The key component of our system is a natural language interface for comparative QA that can be used in personal assistants, chatbots, and similar NLP devices. Comparative QA is a challenging NLP task, since it requires collecting support evidence from many different sources, and direct comparisons of rare objects may be not available even on the entire Web. We take the first step towards a solution for such a task offering a testbed for comparative QA in natural language by probing several methods, making the three best ones available as an online demo.
Text transcripts without punctuation or sentence boundaries are hard to comprehend for both humans and machines. Punctuation marks play a vital role by providing meaning to the sentence and incorrect use or placement of punctuation marks can often alter it. This can impact downstream tasks such as language translation and understanding, pronoun resolution, text summarization, etc. for humans and machines. An automated punctuation restoration (APR) system with minimal human intervention can improve comprehension of text and help users write better. In this paper we describe a multitask modeling approach as a system to restore punctuation in multiple high resource – Germanic (English and German), Romanic (French)– and low resource languages – Indo-Aryan (Hindi) Dravidian (Tamil) – that does not require extensive knowledge of grammar or syntax of a given language for both spoken and written form of text. For German language and the given Indic based languages this is the first towards restoring punctuation and can serve as a baseline for future work.
Conversational Agent for Daily Living Assessment Coaching (CADLAC) is a multi-modal conversational agent system designed to impersonate “individuals” with various levels of ability in activities of daily living (ADLs: e.g., dressing, bathing, mobility, etc.) for use in training professional assessors how to conduct interviews to determine one’s level of functioning. The system is implemented on the MindMeld platform for conversational AI and features a Bidirectional Long Short-Term Memory topic tracker that allows the agent to navigate conversations spanning 18 different ADL domains, a dialogue manager that interfaces with a database of over 10,000 historical ADL assessments, a rule-based Natural Language Generation (NLG) module, and a pre-trained open-domain conversational sub-agent (based on GPT-2) for handling conversation turns outside of the 18 ADL domains. CADLAC is delivered via state-of-the-art web frameworks to handle multiple conversations and users simultaneously and is enabled with voice interface. The paper includes a description of the system design and evaluation of individual components followed by a brief discussion of current limitations and next steps.
Computation-intensive pretrained models have been taking the lead of many natural language processing benchmarks such as GLUE. However, energy efficiency in the process of model training and inference becomes a critical bottleneck. We introduce HULK, a multi-task energy efficiency benchmarking platform for responsible natural language processing. With HULK, we compare pretrained models’ energy efficiency from the perspectives of time and cost. Baseline benchmarking results are provided for further analysis. The fine-tuning efficiency of different pretrained models can differ significantly among different tasks, and fewer parameter number does not necessarily imply better efficiency. We analyzed such a phenomenon and demonstrated the method for comparing the multi-task efficiency of pretrained models. Our platform is available at https://hulkbenchmark.github.io/ .
The problem of estimating the probability distribution of labels has been widely studied as a label distribution learning (LDL) problem, whose applications include age estimation, emotion analysis, and semantic segmentation. We propose a tree-Wasserstein distance regularized LDL algorithm, focusing on hierarchical text classification tasks. We propose predicting the entire label hierarchy using neural networks, where the similarity between predicted and true labels is measured using the tree-Wasserstein distance. Through experiments using synthetic and real-world datasets, we demonstrate that the proposed method successfully considers the structure of labels during training, and it compares favorably with the Sinkhorn algorithm in terms of computation time and memory usage.
With the success of pre-trained language models in recent years, more and more researchers focus on opening the “black box” of these models. Following this interest, we carry out a qualitative and quantitative analysis of constituency grammar in attention heads of BERT and RoBERTa. We employ the syntactic distance method to extract implicit constituency grammar from the attention weights of each head. Our results show that there exist heads that can induce some grammar types much better than baselines, suggesting that some heads act as a proxy for constituency grammar. We also analyze how attention heads’ constituency grammar inducing (CGI) ability changes after fine-tuning with two kinds of tasks, including sentence meaning similarity (SMS) tasks and natural language inference (NLI) tasks. Our results suggest that SMS tasks decrease the average CGI ability of upper layers, while NLI tasks increase it. Lastly, we investigate the connections between CGI ability and natural language understanding ability on QQP and MNLI tasks.
Theories and models of spoken word recognition aim to explain the process of accessing lexical knowledge given an acoustic realization of a word form. There is consensus that phonological and semantic information is crucial for this process. However, there is accumulating evidence that orthographic information could also have an impact on auditory word recognition. This paper presents two models of spoken word recognition that instantiate different hypotheses regarding the influence of orthography on this process. We show that these models reproduce human-like behavior in different ways and provide testable hypotheses for future research on the source of orthographic effects in spoken word recognition.
In this work, we present a methodology that aims at bridging the gap between high and low-resource languages in the context of Open Information Extraction, showcasing it on the Greek language. The goals of this paper are twofold: First, we build Neural Machine Translation (NMT) models for English-to-Greek and Greek-to-English based on the Transformer architecture. Second, we leverage these NMT models to produce English translations of Greek text as input for our NLP pipeline, to which we apply a series of pre-processing and triple extraction tasks. Finally, we back-translate the extracted triples to Greek. We conduct an evaluation of both our NMT and OIE methods on benchmark datasets and demonstrate that our approach outperforms the current state-of-the-art for the Greek natural language.
WikiHow is an open-domain repository of instructional articles for a variety of tasks, which can be revised by users. In this paper, we extract pairwise versions of an instruction before and after a revision was made. Starting from a noisy dataset of revision histories, we specifically extract and analyze edits that involve cases of vagueness in instructions. We further investigate the ability of a neural model to distinguish between two versions of an instruction in our data by adopting a pairwise ranking task from previous work and showing improvements over existing baselines.
Language Identification is the task of identifying a document’s language. For applications like automatic spell checker selection, language identification must use very short strings such as text message fragments. In this work, we reproduce a language identification architecture that Apple briefly sketched in a blog post. We confirm the bi-LSTM model’s performance and find that it outperforms current open-source language identifiers. We further find that its language identification mistakes are due to confusion between related languages.
Institutes are required to catalog their articles with proper subject headings so that the users can easily retrieve relevant articles from the institutional repositories. However, due to the rate of proliferation of the number of articles in these repositories, it is becoming a challenge to manually catalog the newly added articles at the same pace. To address this challenge, we explore the feasibility of automatically annotating articles with Library of Congress Subject Headings (LCSH). We first use web scraping to extract keywords for a collection of articles from the Repository Analytics and Metrics Portal (RAMP). Then, we map these keywords to LCSH names for developing a gold-standard dataset. As a case study, using the subset of Biology-related LCSH concepts, we develop predictive models by formulating this task as a multi-label classification problem. Our experimental results demonstrate the viability of this approach for predicting LCSH for scholarly articles.
While numerous methods have been proposed as defenses against adversarial examples in question answering (QA), these techniques are often model specific, require retraining of the model, and give only marginal improvements in performance over vanilla models. In this work, we present a simple model-agnostic approach to this problem that can be applied directly to any QA model without any retraining. Our method employs an explicit answer candidate reranking mechanism that scores candidate answers on the basis of their content overlap with the question before making the final prediction. Combined with a strong base QAmodel, our method outperforms state-of-the-art defense techniques, calling into question how well these techniques are actually doing and strong these adversarial testbeds are.
Recently, various information retrieval models have been proposed based on pre-trained BERT models, achieving outstanding performance. The majority of such models have been tested on data collections with partial relevance labels, where various potentially relevant documents have not been exposed to the annotators. Therefore, evaluating BERT-based rankers may lead to biased and unfair evaluation results, simply because a relevant document has not been exposed to the annotators while creating the collection. In our work, we aim to better understand a BERT-based ranker’s strengths compared to a BERT-based re-ranker and the initial ranker. To this aim, we investigate BERT-based rankers performance on the Cranfield collection, which comes with full relevance judgment on all documents in the collection. Our results demonstrate the BERT-based full ranker’s effectiveness, as opposed to the BERT-based re-ranker and BM25. Also, analysis shows that there are documents that the BERT-based full-ranker finds that were not found by the initial ranker.
Recommender systems play an important role in e-commerce websites as they improve the customer journey by helping the users find what they want at the right moment. In this paper, we focus on identifying a complementary relationship between the products of an e-commerce company. We propose a content-based recommender system for detecting complementary products, using Siamese Neural Networks (SNN). To this end, we implement and compare two different models: Siamese Convolutional Neural Network (CNN) and Siamese Long Short-Term Memory (LSTM). Moreover, we propose an extension of the SNN approach to handling millions of products in a matter of seconds, and we reduce the training time complexity by half. In the experiments, we show that Siamese LSTM can predict complementary products with an accuracy of ~85% using only the product titles.
We propose a self-supervised method that builds sentence embeddings from the combination of diverse explicit syntactic structures of a sentence. We assume structure is crucial to building consistent representations as we expect sentence meaning to be a function of both syntax and semantic aspects. In this perspective, we hypothesize that some linguistic representations might be better adapted given the considered task or sentence. We, therefore, propose to learn individual representation functions for different syntactic frameworks jointly. Again, by hypothesis, all such functions should encode similar semantic information differently and consequently, be complementary for building better sentential semantic embeddings. To assess such hypothesis, we propose an original contrastive multi-view framework that induces an explicit interaction between models during the training phase. We make experiments combining various structures such as dependency, constituency, or sequential schemes. Our results outperform comparable methods on several tasks from standard sentence embedding benchmarks.
Reasoning about information from multiple parts of a passage to derive an answer is an open challenge for reading-comprehension models. In this paper, we present an approach that reasons about complex questions by decomposing them to simpler subquestions that can take advantage of single-span extraction reading-comprehension models, and derives the final answer according to instructions in a predefined reasoning template. We focus on subtraction based arithmetic questions and evaluate our approach on a subset of the DROP dataset. We show that our approach is competitive with the state of the art while being interpretable and requires little supervision.
The segmentation of emails into functional zones (also dubbed email zoning) is a relevant preprocessing step for most NLP tasks that deal with emails. However, despite the multilingual character of emails and their applications, previous literature regarding email zoning corpora and systems was developed essentially for English. In this paper, we analyse the existing email zoning corpora and propose a new multilingual benchmark composed of 625 emails in Portuguese, Spanish and French. Moreover, we introduce OKAPI, the first multilingual email segmentation model based on a language agnostic sentence encoder. Besides generalizing well for unseen languages, our model is competitive with current English benchmarks, and reached new state-of-the-art performances for domain adaptation tasks in English.
We present a deep neural model of spoken word recognition which is trained to retrieve the meaning of a word (in the form of a word embedding) given its spoken form, a task which resembles that faced by a human listener. Furthermore, we investigate the influence of variability in speech signals on the model’s performance. To this end, we conduct of set of controlled experiments using word-aligned read speech data in German. Our experiments show that (1) the model is more sensitive to dialectical variation than gender variation, and (2) recognition performance of word cognates from related languages reflect the degree of relatedness between languages in our study. Our work highlights the feasibility of modeling human speech perception using deep neural networks.
Sentiment tasks such as hate speech detection and sentiment analysis, especially when performed on languages other than English, are often low-resource. In this study, we exploit the emotional information encoded in emojis to enhance the performance on a variety of sentiment tasks. This is done using a transfer learning approach, where the parameters learned by an emoji-based source task are transferred to a sentiment target task. We analyse the efficacy of the transfer under three conditions, i.e. i) the emoji content and ii) label distribution of the target task as well as iii) the difference between monolingually and multilingually learned source tasks. We find i.a. that the transfer is most beneficial if the target task is balanced with high emoji content. Monolingually learned source tasks have the benefit of taking into account the culturally specific use of emojis and gain up to F1 +0.280 over the baseline.
Neural dependency parsing has achieved remarkable performance for many domains and languages. The bottleneck of massive labelled data limits the effectiveness of these approaches for low resource languages. In this work, we focus on dependency parsing for morphological rich languages (MRLs) in a low-resource setting. Although morphological information is essential for the dependency parsing task, the morphological disambiguation and lack of powerful analyzers pose challenges to get this information for MRLs. To address these challenges, we propose simple auxiliary tasks for pretraining. We perform experiments on 10 MRLs in low-resource settings to measure the efficacy of our proposed pretraining method and observe an average absolute gain of 2 points (UAS) and 3.6 points (LAS).
Almost 30% of the adult population in the world is experiencing or has experience insomnia. Cognitive Behaviour Therapy for insomnia (CBT-I) is one of the most effective treatment, but it has limitations on accessibility and availability. Utilising technology is one of the possible solutions, but existing methods neglect conversational aspects, which plays a critical role in sleep therapy. To address this issue, we propose a PhD project exploring potentials of developing conversational artificial intelligence (AI) for a sleep coaching programme, which is motivated by CBT-I treatment. This PhD project aims to develop natural language processing (NLP) algorithms to allow the system to interact naturally with a user and provide automated analytic system to support human experts. In this paper, we introduce research questions lying under three phases of the sleep coaching programme: triaging, monitoring the progress, and providing coaching. We expect this research project’s outcomes could contribute to the research domains of NLP and AI but also the healthcare field by providing a more accessible and affordable sleep treatment solution and an automated analytic system to lessen the burden of human experts.
Relation extraction is a key task in knowledge extraction, and is commonly defined as the task of identifying relations that hold between entities in text. This thesis proposal addresses the specific task of identifying meta-relations, a higher order family of relations naturally construed as holding between other relations which includes temporal, comparative, and causal relations. More specifically, we aim to develop theoretical underpinnings and practical solutions for the challenges of (1) incorporating meta-relations into conceptualisations and annotation schemes for (lower-order) relations and named entities, (2) obtaining annotations for them with tolerable cognitive load on annotators, (3) creating models capable of reliably extracting meta-relations, and related to that (4) addressing the limited-data problem exacerbated by the introduction of meta-relations into the learning task. We explore recent works in relation extraction and discuss our plans to formally conceptualise meta-relations for the domain of user-generated health texts, and create a new dataset, annotation scheme and models for meta-relation extraction.
State-of-the-art (SOTA) neural machine translation (NMT) systems translate texts at sentence level, ignoring context: intra-textual information, like the previous sentence, and extra-textual information, like the gender of the speaker. As a result, some sentences are translated incorrectly. Personalised NMT (PersNMT) and document-level NMT (DocNMT) incorporate this information into the translation process. Both fields are relatively new and previous work within them is limited. Moreover, there are no readily available robust evaluation metrics for them, which makes it difficult to develop better systems, as well as track global progress and compare different methods. This thesis proposal focuses on PersNMT and DocNMT for the domain of dialogue extracted from TV subtitles in five languages: English, Brazilian Portuguese, German, French and Polish. Three main challenges are addressed: (1) incorporating extra-textual information directly into NMT systems; (2) improving the machine translation of cohesion devices; (3) reliable evaluation for PersNMT and DocNMT.
Most works in food computing focus on generating new recipes from scratch. However, there is a large number of new online recipes generated daily with a large number of users reviews, with recommendations to improve the recipe flavor and ideas to modify them. This fact encourages the use of these data for obtaining improved and customized versions. In this thesis, we propose an adaptation engine based on fine-tuning a word embedding model. We will capture, in an unsupervised way, the semantic meaning of the recipe ingredients. We will use their word embedding representations to align them to external databases, thus enriching their data. The adaptation engine will use this food data to modify a recipe into another fitting specific user preferences (e.g., decrease caloric intake or make a recipe). We plan to explore different types of recipe adaptations while preserving recipe essential features such as cuisine style and essence simultaneously. We will also modify the rest of the recipe to the new changes to be reproducible.
We propose the Tough Mentions Recall (TMR) metrics to supplement traditional named entity recognition (NER) evaluation by examining recall on specific subsets of ”tough” mentions: unseen mentions, those whose tokens or token/type combination were not observed in training, and type-confusable mentions, token sequences with multiple entity types in the test data. We demonstrate the usefulness of these metrics by evaluating corpora of English, Spanish, and Dutch using five recent neural architectures. We identify subtle differences between the performance of BERT and Flair on two English NER corpora and identify a weak spot in the performance of current models in Spanish. We conclude that the TMR metrics enable differentiation between otherwise similar-scoring systems and identification of patterns in performance that would go unnoticed from overall precision, recall, and F1.
This paper evaluates the performance of several modern subword segmentation methods in a low-resource neural machine translation setting. We compare segmentations produced by applying BPE at the token or sentence level with morphologically-based segmentations from LMVR and MORSEL. We evaluate translation tasks between English and each of Nepali, Sinhala, and Kazakh, and predict that using morphologically-based segmentation methods would lead to better performance in this setting. However, comparing to BPE, we find that no consistent and reliable differences emerge between the segmentation methods. While morphologically-based methods outperform BPE in a few cases, what performs best tends to vary across tasks, and the performance of segmentation methods is often statistically indistinguishable.
Generating diverse texts is an important factor for unsupervised text generation. One approach is to produce the diversity of texts conditioned by the sampled latent code. Although several generative adversarial networks (GANs) have been proposed thus far, these models still suffer from mode-collapsing if the models are not pre-trained. In this paper, we propose a GAN model that aims to improve the approach to generating diverse texts conditioned by the latent space. The generator of our model uses Gumbel-Softmax distribution for the word sampling process. To ensure that the text is generated conditioned upon the sampled latent code, reconstruction loss is introduced in our objective function. The discriminator of our model iteratively inspects incomplete partial texts and learns to distinguish whether they are real or fake by using the standard GAN objective function. Experimental results using the COCO Image Captions dataset show that, although our model is not pre-trained, the performance of our model is quite competitive with the existing baseline models, which requires pre-training.
We explore cross-lingual transfer of register classification for web documents. Registers, that is, text varieties such as blogs or news are one of the primary predictors of linguistic variation and thus affect the automatic processing of language. We introduce two new register-annotated corpora, FreCORE and SweCORE, for French and Swedish. We demonstrate that deep pre-trained language models perform strongly in these languages and outperform previous state-of-the-art in English and Finnish. Specifically, we show 1) that zero-shot cross-lingual transfer from the large English CORE corpus can match or surpass previously published monolingual models, and 2) that lightweight monolingual classification requiring very little training data can reach or surpass our zero-shot performance. We further analyse classification results finding that certain registers continue to pose challenges in particular for cross-lingual transfer.
Type- and token-based embedding architectures are still competing in lexical semantic change detection. The recent success of type-based models in SemEval-2020 Task 1 has raised the question why the success of token-based models on a variety of other NLP tasks does not translate to our field. We investigate the influence of a range of variables on clusterings of BERT vectors and show that its low performance is largely due to orthographic information on the target word, which is encoded even in the higher layers of BERT representations. By reducing the influence of orthography we considerably improve BERT’s performance.
The present paper investigates the impact of the anaphoric one words in English on the Neural Machine Translation (NMT) process using English-Hindi as source and target language pair. As expected, the experimental results show that the state-of-the-art Google English-Hindi NMT system achieves significantly poorly on sentences containing anaphoric ones as compared to the sentences containing regular, non-anaphoric ones. But, more importantly, we note that amongst the anaphoric words, the noun class is clearly much harder for NMT than the determinatives. This reaffirms the linguistic disparity of the two phenomenon in recent theoretical syntactic literature, despite the obvious surface similarities.
Unsupervised parsing learns a syntactic parser from training sentences without parse tree annotations. Recently, there has been a resurgence of interest in unsupervised parsing, which can be attributed to the combination of two trends in the NLP community: a general trend towards unsupervised training or pre-training, and an emerging trend towards finding or modeling linguistic structures in neural models. In this tutorial, we will introduce to the general audience what unsupervised parsing does and how it can be useful for and beyond syntactic parsing. We will then provide a systematic overview of major classes of approaches to unsupervised parsing, namely generative and discriminative approaches, and analyze their relative strengths and weaknesses. We will cover both decade-old statistical approaches and more recent neural approaches to give the audience a sense of the historical and recent development of the field. We will also discuss emerging research topics such as BERT-based approaches and visually grounded learning.
The success of NLP research is founded on high-quality annotated datasets, which are usually obtained from multiple expert annotators or crowd workers. The standard practice to training machine learning models is to first adjudicate the disagreements and then perform the training. To this end, there has been a lot of work on aggregating annotations, particularly for classification tasks. However, many other tasks, particularly in NLP, have unique characteristics not considered by standard models of annotation, e.g., label interdependencies in sequence labelling tasks, unrestricted labels for anaphoric annotation, or preference labels for ranking texts. In recent years, researchers have picked up on this and are covering the gap. A first objective of this tutorial is to connect NLP researchers with state-of-the-art aggregation models for a diverse set of canonical language annotation tasks. There is also a growing body of recent work arguing that following the convention and training with adjudicated labels ignores any uncertainty the labellers had in their classifications, which results in models with poorer generalisation capabilities. Therefore, a second objective of this tutorial is to teach NLP workers how they can augment their (deep) neural models to learn from data with multiple interpretations.
Speech translation is the translation of speech in one language typically to text in another, traditionally accomplished through a combination of automatic speech recognition and machine translation. Speech translation has attracted interest for many years, but the recent successful applications of deep learning to both individual tasks have enabled new opportunities through joint modeling, in what we today call ‘end-to-end speech translation.’ In this tutorial we will introduce the techniques used in cutting-edge research on speech translation. Starting from the traditional cascaded approach, we will given an overview on data sources and model architectures to achieve state-of-the art performance with end-to-end speech translation for both high- and low-resource languages. In addition, we will discuss methods to evaluate analyze the proposed solutions, as well as the challenges faced when applying speech translation models for real-world applications.
The reviewing procedure has been identified as one of the major issues in the current situation of the NLP field. While it is implicitly assumed that junior researcher learn reviewing during their PhD project, this might not always be the case. Additionally, with the growing NLP community and the efforts in the context of widening the NLP community, researchers joining the field might not have the opportunity to practise reviewing. This tutorial fills in this gap by providing an opportunity to learn the basics of reviewing. Also more experienced researchers might find this tutorial interesting to revise their reviewing procedure.
Unsupervised cross-lingual language representation initialization methods, together with mechanisms such as denoising and back-translation, have advanced unsupervised neural machine translation (UNMT), which has achieved impressive results. Meanwhile, there are still several challenges for UNMT. This tutorial first introduces the background and the latest progress of UNMT. We then examine a number of challenges to UNMT and give empirical results on how well the technology currently holds up.
When tackling a task in a given domain, it has been shown that adapting a model to the domain using raw text data before training on the supervised task improves performance versus solely training on the task. The downside is that a lot of domain data is required and if we want to tackle tasks in n domains, we require n models each adapted on domain data before task learning. Storing and using these models separately can be prohibitive for low-end devices. In this paper we show that domain adaptation can be generalised to cover multiple domains. Specifically, a single model can be trained across various domains at the same time with minimal drop in performance, even when we use less data and resources. Thus, instead of training multiple models, we can train a single multidomain model saving on computational resources and training time.
Contextual embedding models such as BERT can be easily fine-tuned on labeled samples to create a state-of-the-art model for many downstream tasks. However, the fine-tuned BERT model suffers considerably from unlabeled data when applied to a different domain. In unsupervised domain adaptation, we aim to train a model that works well on a target domain when provided with labeled source samples and unlabeled target samples. In this paper, we propose a pseudo-label guided method for unsupervised domain adaptation. Two models are fine-tuned on labeled source samples as pseudo labeling models. To learn representations for the target domain, one of those models is adapted by masked language modeling from the target domain. Then those models are used to assign pseudo-labels to target samples. We train the final model with those samples. We evaluate our method on named entity segmentation and sentiment analysis tasks. These experiments show that our approach outperforms baseline methods.
In this paper, we propose conditional adversarial networks (CANs), a framework that explores the relationship between the shared features and the label predictions to impose stronger discriminability to the learned features, for multi-domain text classification (MDTC). The proposed CAN introduces a conditional domain discriminator to model the domain variance in both the shared feature representations and the class-aware information simultaneously, and adopts entropy conditioning to guarantee the transferability of the shared features. We provide theoretical analysis for the CAN framework, showing that CAN’s objective is equivalent to minimizing the total divergence among multiple joint distributions of shared features and label predictions. Therefore, CAN is a theoretically sound adversarial network that discriminates over multiple distributions. Evaluation results on two MDTC benchmarks show that CAN outperforms prior methods. Further experiments demonstrate that CAN has a good ability to generalize learned knowledge to unseen domains.
This paper provides the first experimental study on the impact of using domain-specific representations on a BERT-based multi-task spoken language understanding (SLU) model for multi-domain applications. Our results on a real-world dataset covering three languages indicate that by using domain-specific representations learned adversarially, model performance can be improved across all of the three SLU subtasks domain classification, intent classification and slot filling. Gains are particularly large for domains with limited training data.
Social media such as Twitter provide valuable information to crisis managers and affected people during natural disasters. Machine learning can help structure and extract information from the large volume of messages shared during a crisis; however, the constantly evolving nature of crises makes effective domain adaptation essential. Supervised classification is limited by unchangeable class labels that may not be relevant to new events, and unsupervised topic modelling by insufficient prior knowledge. In this paper, we bridge the gap between the two and show that BERT embeddings finetuned on crisis-related tweet classification can effectively be used to adapt to a new crisis, discovering novel topics while preserving relevant classes from supervised training, and leveraging bidirectional self-attention to extract topic keywords. We create a dataset of tweets from a snowstorm to evaluate our method’s transferability to new crises, and find that it outperforms traditional topic models in both automatic, and human evaluations grounded in the needs of crisis managers. More broadly, our method can be used for textual domain adaptation where the latent classes are unknown but overlap with known classes from other domains.
While high performance have been obtained for high-resource languages, performance on low-resource languages lags behind. In this paper we focus on the parsing of the low-resource language Frisian. We use a sample of code-switched, spontaneously spoken data, which proves to be a challenging setup. We propose to train a parser specifically tailored towards the target domain, by selecting instances from multiple treebanks. Specifically, we use Latent Dirichlet Allocation (LDA), with word and character N-grams. We use a deep biaffine parser initialized with mBERT. The best single source treebank (nl_alpino) resulted in an LAS of 54.7 whereas our data selection outperformed the single best transfer treebank and led to 55.6 LAS on the test data. Additional experiments consisted of removing diacritics from our Frisian data, creating more similar training data by cropping sentences and running our best model using XLM-R. These experiments did not lead to a better performance.
Genre and domain are often used interchangeably, but are two different properties of a text. Successful parser adaptation requires both cross-domain and cross-genre sensitivity (Rehbein and Bildhauer, 2017). While the impact domain differences have on parser performance degradation is more easily measurable in respect to lexical differences, impact of genre differences can be more nuanced. With the predominance of pre-trained language models (LMs; e.g. BERT (Devlin et al., 2019)), there are now additional complexities in developing cross-genre sensitive models due to the infusion of linguistic characteristics derived from, usually, a third genre. We perform a systematic set of experiments using two neural constituency parsers to examine how different parsers behave in combination with different BERT models with varying source and target genres in English and Swedish. We find that there is extensive difficulty in predicting the best source due to the complex interactions between genres, parsers, and LMs. Additionally, the influence of the data used to derive the underlying BERT model heavily influences how best to create more robust and effective cross-genre parsing models.
In meta-learning, the knowledge learned from previous tasks is transferred to new ones, but this transfer only works if tasks are related. Sharing information between unrelated tasks might hurt performance, and it is unclear how to transfer knowledge across tasks that have a hierarchical structure. Our research extends a meta-learning model, MAML, by exploiting hierarchical task relationships. Our algorithm, TreeMAML, adapts the model to each task with a few gradient steps, but the adaptation follows the hierarchical tree structure: in each step, gradients are pooled across tasks clusters and subsequent steps follow down the tree. We also implement a clustering algorithm that generates the tasks tree without previous knowledge of the task structure, allowing us to make use of implicit relationships between the tasks. We show that TreeMAML successfully trains natural language processing models for cross-lingual Natural Language Inference by taking advantage of the language phylogenetic tree. This result is useful since most languages in the world are under-resourced and the improvement on cross-lingual transfer allows the internationalization of NLP models.
Achieving satisfying performance in machine translation on domains for which there is no training data is challenging. Traditional supervised domain adaptation is not suitable for addressing such zero-resource domains because it relies on in-domain parallel data. We show that when in-domain parallel data is not available, access to document-level context enables better capturing of domain generalities compared to only having access to a single sentence. Having access to more information provides a more reliable domain estimation. We present two document-level Transformer models which are capable of using large context sizes and we compare these models against strong Transformer baselines. We obtain improvements for the two zero-resource domains we study. We additionally provide an analysis where we vary the amount of context and look at the case where in-domain data is available.
Retrieval question answering (ReQA) is the task of retrieving a sentence-level answer to a question from an open corpus (Ahmad et al.,2019).This dataset paper presents MultiReQA, a new multi-domain ReQA evaluation suite composed of eight retrieval QA tasks drawn from publicly available QA datasets. We explore systematic retrieval based evaluation and transfer learning across domains over these datasets using a number of strong base-lines including two supervised neural models, based on fine-tuning BERT and USE-QA models respectively, as well as a surprisingly effective information retrieval baseline, BM25. Five of these tasks contain both training and test data, while three contain test data only. Performing cross training on the five tasks with training data shows that while a general model covering all domains is achievable, the best performance is often obtained by training exclusively on in-domain data.
Advances in transfer learning and domain adaptation have raised hopes that once-challenging NLP tasks are ready to be put to use for sophisticated information extraction needs. In this work, we describe an effort to do just that – combining state-of-the-art neural methods for negation detection, document time relation extraction, and aspectual link prediction, with the eventual goal of extracting drug timelines from electronic health record text. We train on the THYME colon cancer corpus and test on both the THYME brain cancer corpus and an internal corpus, and show that performance of the combined systems is unacceptable despite good performance of individual systems. Although domain adaptation shows improvements on each individual system, the model selection problem is a barrier to improving overall pipeline performance.
Models such as mBERT and XLMR have shown success in solving Code-Mixed NLP tasks even though they were not exposed to such text during pretraining. Code-Mixed NLP models have relied on using synthetically generated data along with naturally occurring data to improve their performance. Finetuning mBERT on such data improves it’s code-mixed performance, but the benefits of using the different types of Code-Mixed data aren’t clear. In this paper, we study the impact of finetuning with different types of code-mixed data and outline the changes that occur to the model during such finetuning. Our findings suggest that using naturally occurring code-mixed data brings in the best performance improvement after finetuning and that finetuning with any type of code-mixed text improves the responsivity of it’s attention heads to code-mixed text inputs.