Yin Zhang


ERNIE-Layout: Layout Knowledge Enhanced Pre-training for Visually-rich Document Understanding
Qiming Peng | Yinxu Pan | Wenjin Wang | Bin Luo | Zhenyu Zhang | Zhengjie Huang | Yuhui Cao | Weichong Yin | Yongfeng Chen | Yin Zhang | Shikun Feng | Yu Sun | Hao Tian | Hua Wu | Haifeng Wang
Findings of the Association for Computational Linguistics: EMNLP 2022

Recent years have witnessed the rise and success of pre-training techniques in visually-rich document understanding. However, most existing methods lack the systematic mining and utilization of layout-centered knowledge, leading to sub-optimal performances. In this paper, we propose ERNIE-Layout, a novel document pre-training solution with layout knowledge enhancement in the whole workflow, to learn better representations that combine the features from text, layout, and image. Specifically, we first rearrange input sequences in the serialization stage, and then present a correlative pre-training task, reading order prediction, to learn the proper reading order of documents. To improve the layout awareness of the model, we integrate a spatial-aware disentangled attention into the multi-modal transformer and a replaced regions prediction task into the pre-training phase. Experimental results show that ERNIE-Layout achieves superior performance on various downstream tasks, setting new state-of-the-art on key information extraction, document image classification, and document question answering datasets. The code and models are publicly available at PaddleNLP.

Continual Few-shot Intent Detection
Guodun Li | Yuchen Zhai | Qianglong Chen | Xing Gao | Ji Zhang | Yin Zhang
Proceedings of the 29th International Conference on Computational Linguistics

Intent detection is at the core of task-oriented dialogue systems. Existing intent detection systems are typically trained with a large amount of data over a predefined set of intent classes. However, newly emerged intents in multiple domains are commonplace in the real world. And it is time-consuming and impractical for dialogue systems to re-collect enough annotated data and re-train the model. These limitations call for an intent detection system that could continually recognize new intents with very few labeled examples. In this work, we study the Continual Few-shot Intent Detection (CFID) problem and construct a benchmark consisting of nine tasks with multiple domains and imbalanced classes. To address the key challenges of (a) catastrophic forgetting during continuous learning and (b) negative knowledge transfer across tasks, we propose the Prefix-guided Lightweight Encoder (PLE) with three auxiliary strategies, namely Pseudo Samples Replay (PSR), Teacher Knowledge Transfer (TKT) and Dynamic Weighting Replay (DWR). Extensive experiments demonstrate the effectiveness and efficiency of our method in preventing catastrophic forgetting and encouraging positive knowledge transfer across tasks.

OTSeq2Set: An Optimal Transport Enhanced Sequence-to-Set Model for Extreme Multi-label Text Classification
Jie Cao | Yin Zhang
Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

Extreme multi-label text classification (XMTC) is the task of finding the most relevant subset labels from an extremely large-scale label collection. Recently, some deep learning models have achieved state-of-the-art results in XMTC tasks. These models commonly predict scores for all labels by a fully connected layer as the last layer of the model. However, such models can’t predict a relatively complete and variable-length label subset for each document, because they select positive labels relevant to the document by a fixed threshold or take top k labels in descending order of scores. A less popular type of deep learning models called sequence-to-sequence (Seq2Seq) focus on predicting variable-length positive labels in sequence style. However, the labels in XMTC tasks are essentially an unordered set rather than an ordered sequence, the default order of labels restrains Seq2Seq models in training. To address this limitation in Seq2Seq, we propose an autoregressive sequence-to-set model for XMTC tasks named OTSeq2Set. Our model generates predictions in student-forcing scheme and is trained by a loss function based on bipartite matching which enables permutation-invariance. Meanwhile, we use the optimal transport distance as a measurement to force the model to focus on the closest labels in semantic label space. Experiments show that OTSeq2Set outperforms other competitive baselines on 4 benchmark datasets. Especially, on the Wikipedia dataset with 31k labels, it outperforms the state-of-the-art Seq2Seq method by 16.34% in micro-F1 score. The code is available at https://github.com/caojie54/OTSeq2Set.


KACE: Generating Knowledge Aware Contrastive Explanations for Natural Language Inference
Qianglong Chen | Feng Ji | Xiangji Zeng | Feng-Lin Li | Ji Zhang | Haiqing Chen | Yin Zhang
Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers)

In order to better understand the reason behind model behaviors (i.e., making predictions), most recent works have exploited generative models to provide complementary explanations. However, existing approaches in NLP mainly focus on “WHY A” rather than contrastive “WHY A NOT B”, which is shown to be able to better distinguish confusing candidates and improve data efficiency in other research fields.In this paper, we focus on generating contrastive explanations with counterfactual examples in NLI and propose a novel Knowledge-Aware Contrastive Explanation generation framework (KACE).Specifically, we first identify rationales (i.e., key phrases) from input sentences, and use them as key perturbations for generating counterfactual examples. After obtaining qualified counterfactual examples, we take them along with original examples and external knowledge as input, and employ a knowledge-aware generative pre-trained language model to generate contrastive explanations. Experimental results show that contrastive explanations are beneficial to fit the scenarios by clarifying the difference between the predicted answer and other possible wrong ones. Moreover, we train an NLI model enhanced with contrastive explanations and achieves an accuracy of 91.9% on SNLI, gaining improvements of 5.7% against ETPA (“Explain-Then-Predict-Attention”) and 0.6% against NILE (“WHY A”).

Disentangled Code Representation Learning for Multiple Programming Languages
Jingfeng Zhang | Haiwen Hong | Yin Zhang | Yao Wan | Ye Liu | Yulei Sui
Findings of the Association for Computational Linguistics: ACL-IJCNLP 2021

UnClE: Explicitly Leveraging Semantic Similarity to Reduce the Parameters of Word Embeddings
Zhi Li | Yuchen Zhai | Chengyu Wang | Minghui Qiu | Kailiang Li | Yin Zhang
Findings of the Association for Computational Linguistics: EMNLP 2021

Natural language processing (NLP) models often require a massive number of parameters for word embeddings, which limits their application on mobile devices. Researchers have employed many approaches, e.g. adaptive inputs, to reduce the parameters of word embeddings. However, existing methods rarely pay attention to semantic information. In this paper, we propose a novel method called Unique and Class Embeddings (UnClE), which explicitly leverages semantic similarity with weight sharing to reduce the dimensionality of word embeddings. Inspired by the fact that words with similar semantic can share a part of weights, we divide the embeddings of words into two parts: unique embedding and class embedding. The former is one-to-one mapping like traditional embedding, while the latter is many-to-one mapping and learn the representation of class information. Our method is suitable for both word-level and sub-word level models and can be used to reduce both input and output embeddings. Experimental results on the standard WMT 2014 English-German dataset show that our method is able to reduce the parameters of word embeddings by more than 11x, with about 93% performance retaining in BLEU metrics. For language modeling task, our model can reduce word embeddings by 6x or 11x on PTB/WT2 dataset at the cost of a certain degree of performance degradation.

Fix-Filter-Fix: Intuitively Connect Any Models for Effective Bug Fixing
Haiwen Hong | Jingfeng Zhang | Yin Zhang | Yao Wan | Yulei Sui
Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing

Locating and fixing bugs is a time-consuming task. Most neural machine translation (NMT) based approaches for automatically bug fixing lack generality and do not make full use of the rich information in the source code. In NMT-based bug fixing, we find some predicted code identical to the input buggy code (called unchanged fix) in NMT-based approaches due to high similarity between buggy and fixed code (e.g., the difference may only appear in one particular line). Obviously, unchanged fix is not the correct fix because it is the same as the buggy code that needs to be fixed. Based on these, we propose an intuitive yet effective general framework (called Fix-Filter-Fix or Fˆ3) for bug fixing. Fˆ3 connects models with our filter mechanism to filter out the last model’s unchanged fix to the next. We propose an Fˆ3 theory that can quantitatively and accurately calculate the Fˆ3 lifting effect. To evaluate, we implement the Seq2Seq Transformer (ST) and the AST2Seq Transformer (AT) to form some basic Fˆ3 instances, called Fˆ3_ST+AT and Fˆ3_AT+ST. Comparing them with single model approaches and many model connection baselines across four datasets validates the effectiveness and generality of Fˆ3 and corroborates our findings and methodology.

Meta Distant Transfer Learning for Pre-trained Language Models
Chengyu Wang | Haojie Pan | Minghui Qiu | Jun Huang | Fei Yang | Yin Zhang
Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing

With the wide availability of Pre-trained Language Models (PLMs), multi-task fine-tuning across domains has been extensively applied. For tasks related to distant domains with different class label sets, PLMs may memorize non-transferable knowledge for the target domain and suffer from negative transfer. Inspired by meta-learning, we propose the Meta Distant Transfer Learning (Meta-DTL) framework to learn the cross-task knowledge for PLM-based methods. Meta-DTL first employs task representation learning to mine implicit relations among multiple tasks and classes. Based on the results, it trains a PLM-based meta-learner to capture the transferable knowledge across tasks. The weighted maximum entropy regularizers are proposed to make meta-learner more task-agnostic and unbiased. Finally, the meta-learner can be fine-tuned to fit each task with better parameter initialization. We evaluate Meta-DTL using both BERT and ALBERT on seven public datasets. Experiment results confirm the superiority of Meta-DTL as it consistently outperforms strong baselines. We find that Meta-DTL is highly effective when very few data is available for the target task.

Dialogue State Tracking with Multi-Level Fusion of Predicted Dialogue States and Conversations
Jingyao Zhou | Haipang Wu | Zehao Lin | Guodun Li | Yin Zhang
Proceedings of the 22nd Annual Meeting of the Special Interest Group on Discourse and Dialogue

Most recently proposed approaches in dialogue state tracking (DST) leverage the context and the last dialogue states to track current dialogue states, which are often slot-value pairs. Although the context contains the complete dialogue information, the information is usually indirect and even requires reasoning to obtain. The information in the lastly predicted dialogue states is direct, but when there is a prediction error, the dialogue information from this source will be incomplete or erroneous. In this paper, we propose the Dialogue State Tracking with Multi-Level Fusion of Predicted Dialogue States and Conversations network (FPDSC). This model extracts information of each dialogue turn by modeling interactions among each turn utterance, the corresponding last dialogue states, and dialogue slots. Then the representation of each dialogue turn is aggregated by a hierarchical structure to form the passage information, which is utilized in the current turn of DST. Experimental results validate the effectiveness of the fusion network with 55.03% and 59.07% joint accuracy on MultiWOZ 2.0 and MultiWOZ 2.1 datasets, which reaches the state-of-the-art performance. Furthermore, we conduct the deleted-value and related-slot experiments on MultiWOZ 2.1 to evaluate our model.


Improving Commonsense Question Answering by Graph-based Iterative Retrieval over Multiple Knowledge Sources
Qianglong Chen | Feng Ji | Haiqing Chen | Yin Zhang
Proceedings of the 28th International Conference on Computational Linguistics

In order to facilitate natural language understanding, the key is to engage commonsense or background knowledge. However, how to engage commonsense effectively in question answering systems is still under exploration in both research academia and industry. In this paper, we propose a novel question-answering method by integrating multiple knowledge sources, i.e. ConceptNet, Wikipedia, and the Cambridge Dictionary, to boost the performance. More concretely, we first introduce a novel graph-based iterative knowledge retrieval module, which iteratively retrieves concepts and entities related to the given question and its choices from multiple knowledge sources. Afterward, we use a pre-trained language model to encode the question, retrieved knowledge and choices, and propose an answer choice-aware attention mechanism to fuse all hidden representations of the previous modules. Finally, the linear classifier for specific tasks is used to predict the answer. Experimental results on the CommonsenseQA dataset show that our method significantly outperforms other competitive methods and achieves the new state-of-the-art. In addition, further ablation studies demonstrate the effectiveness of our graph-based iterative knowledge retrieval module and the answer choice-aware attention module in retrieving and synthesizing background knowledge from multiple knowledge sources.

Infusing Disease Knowledge into BERT for Health Question Answering, Medical Inference and Disease Name Recognition
Yun He | Ziwei Zhu | Yin Zhang | Qin Chen | James Caverlee
Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP)

Knowledge of a disease includes information of various aspects of the disease, such as signs and symptoms, diagnosis and treatment. This disease knowledge is critical for many health-related and biomedical tasks, including consumer health question answering, medical language inference and disease name recognition. While pre-trained language models like BERT have shown success in capturing syntactic, semantic, and world knowledge from text, we find they can be further complemented by specific information like knowledge of symptoms, diagnoses, treatments, and other disease aspects. Hence, we integrate BERT with disease knowledge for improving these important tasks. Specifically, we propose a new disease knowledge infusion training procedure and evaluate it on a suite of BERT models including BERT, BioBERT, SciBERT, ClinicalBERT, BlueBERT, and ALBERT. Experiments over the three tasks show that these models can be enhanced in nearly all cases, demonstrating the viability of disease knowledge infusion. For example, accuracy of BioBERT on consumer health question answering is improved from 68.29% to 72.09%, while new SOTA results are observed in two datasets. We make our data and code freely available.

Counterfactual Generator: A Weakly-Supervised Method for Named Entity Recognition
Xiangji Zeng | Yunliang Li | Yuchen Zhai | Yin Zhang
Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP)

Past progress on neural models has proven that named entity recognition is no longer a problem if we have enough labeled data. However, collecting enough data and annotating them are labor-intensive, time-consuming, and expensive. In this paper, we decompose the sentence into two parts: entity and context, and rethink the relationship between them and model performance from a causal perspective. Based on this, we propose the Counterfactual Generator, which generates counterfactual examples by the interventions on the existing observational examples to enhance the original dataset. Experiments across three datasets show that our method improves the generalization ability of models under limited observational examples. Besides, we provide a theoretical foundation by using a structural causal model to explore the spurious correlations between input features and output labels. We investigate the causal effects of entity or context on model performance under both conditions: the non-augmented and the augmented. Interestingly, we find that the non-spurious correlations are more located in entity representation rather than context representation. As a result, our method eliminates part of the spurious correlations between context representation and output labels. The code is available at https://github.com/xijiz/cfgen.

PARADE: A New Dataset for Paraphrase Identification Requiring Computer Science Domain Knowledge
Yun He | Zhuoer Wang | Yin Zhang | Ruihong Huang | James Caverlee
Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP)

We present a new benchmark dataset called PARADE for paraphrase identification that requires specialized domain knowledge. PARADE contains paraphrases that overlap very little at the lexical and syntactic level but are semantically equivalent based on computer science domain knowledge, as well as non-paraphrases that overlap greatly at the lexical and syntactic level but are not semantically equivalent based on this domain knowledge. Experiments show that both state-of-the-art neural models and non-expert human annotators have poor performance on PARADE. For example, BERT after fine-tuning achieves an F1 score of 0.709, which is much lower than its performance on other paraphrase identification datasets. PARADE can serve as a resource for researchers interested in testing models that incorporate domain knowledge. We make our data and code freely available.


Task-Oriented Conversation Generation Using Heterogeneous Memory Networks
Zehao Lin | Xinjing Huang | Feng Ji | Haiqing Chen | Yin Zhang
Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP)

How to incorporate external knowledge into a neural dialogue model is critically important for dialogue systems to behave like real humans. To handle this problem, memory networks are usually a great choice and a promising way. However, existing memory networks do not perform well when leveraging heterogeneous information from different sources. In this paper, we propose a novel and versatile external memory networks called Heterogeneous Memory Networks (HMNs), to simultaneously utilize user utterances, dialogue history and background knowledge tuples. In our method, historical sequential dialogues are encoded and stored into the context-aware memory enhanced by gating mechanism while grounding knowledge tuples are encoded and stored into the context-free memory. During decoding, the decoder augmented with HMNs recurrently selects each word in one response utterance from these two memories and a general vocabulary. Experimental results on multiple real-world datasets show that HMNs significantly outperform the state-of-the-art data-driven task-oriented dialogue models in most domains.