publications

2024

1. ACL
   2024

   Harder Task Needs More Experts: Dynamic Routing in MoE Models

   Quzhe Huang, Zhenwei An, Nan Zhuang, and 8 more authors

   In Proceedings of the Association for Computational Linguistics: ACL 2024, Aug 2024
2. ACL
   2024

   Teaching Large Language Models an Unseen Language on the Fly

   Chen Zhang, Xiao Liu, Jiuheng Lin, and 1 more author

   In Findings of the Association for Computational Linguistics: ACL 2024, Aug 2024
3. ACL
   2024

   Are LLMs Capable of Data-based Statistical and Causal Reasoning? Benchmarking Advanced Quantitative Reasoning with Data

   Xiao Liu, Zirui Wu, Xueqing Wu, and 3 more authors

   In Findings of the Association for Computational Linguistics: ACL 2024, Aug 2024
4. ACL
   2024

   ELLA: Empowering LLMs for Interpretable, Accurate and Informative Legal Advice

   Yutong Hu, Kangcheng Luo, and Yansong Feng

   In Proceedings of the Association for Computational Linguistics: System Demonstration, Aug 2024
5. ACL
   2024

   MC^2: Towards Transparent and Culturally-Aware NLP for Minority Languages in China

   Chen Zhang, Mingxu Tao, Quzhe Huang, and 3 more authors

   In Proceedings of the Association for Computational Linguistics: ACL 2024, Aug 2024
6. LREC-COLING
   2024

   Motion Generation from Fine-grained Textual Descriptions

   Kunhang Li, and Yansong Feng

   In the Joint International Conference on Computational Linguistics, Language Resources and Evaluation, Aug 2024

   Abs arXiv PDF Video Poster Slides Website

   The task of text2motion is to generate human motion sequences from given textual descriptions, where the model explores diverse mappings from natural language instructions to human body movements. While most existing works are confined to coarse-grained motion descriptions, e.g., “A man squats.", fine-grained descriptions specifying movements of relevant body parts are barely explored. Models trained with coarse-grained texts may not be able to learn mappings from fine-grained motion-related words to motion primitives, resulting in the failure to generate motions from unseen descriptions. In this paper, we build a large-scale language-motion dataset specializing in fine-grained textual descriptions, FineHumanML3D, by feeding GPT-3.5-turbo with step-by-step instructions with pseudo-code compulsory checks. Accordingly, we design a new text2motion model, FineMotionDiffuse, making full use of fine-grained textual information. Our quantitative evaluation shows that FineMotionDiffuse trained on FineHumanML3D improves FID by a large margin of 0.38, compared with competitive baselines. According to the qualitative evaluation and case study, our model outperforms MotionDiffuse in generating spatially or chronologically composite motions, by learning the implicit mappings from fine-grained descriptions to the corresponding basic motions.

7. LREC-COLING
   2024

   Probing Multimodal Large Language Models for Global and Local Semantic Representations

   Mingxu Tao, Quzhe Huang, Kun Xu, and 3 more authors

   In Proceedings of the 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation (LREC-COLING 2024), May 2024

   Abs

   The advancement of Multimodal Large Language Models (MLLMs) has greatly accelerated the development of applications in understanding integrated texts and images. Recent works leverage image-caption datasets to train MLLMs, achieving state-of-the-art performance on image-to-text tasks. However, there are few studies exploring which layers of MLLMs make the most effort to the global image information, which plays vital roles in multimodal comprehension and generation. In this study, we find that the intermediate layers of models can encode more global semantic information, whose representation vectors perform better on visual-language entailment tasks, rather than the topmost layers. We further probe models regarding local semantic representations through object recognition tasks. We find that the topmost layers may excessively focus on local information, leading to a diminished ability to encode global information. Our code and data are released via https://github.com/kobayashikanna01/probing_MLLM_rep.

8. NAACL-HLT
   2024

   CASA: Causality-driven Argument Sufficiency Assessment

   Xiao Liu, Yansong Feng, and Kai-Wei Chang

   In Proceedings of the 2024 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers), Jun 2024

   Abs

   The argument sufficiency assessment task aims to determine if the premises of a given argument support its conclusion.To tackle this task, existing works often train a classifier on data annotated by humans. However, annotating data is laborious, and annotations are often inconsistent due to subjective criteria. Motivated by the definition of probability of sufficiency (PS) in the causal literature, we proposeCASA, a zero-shot causality-driven argument sufficiency assessment framework. PS measures how likely introducing the premise event would lead to the conclusion when both the premise and conclusion events are absent. To estimate this probability, we propose to use large language models (LLMs) to generate contexts that are inconsistent with the premise and conclusion and revise them by injecting the premise event.Experiments on two logical fallacy detection datasets demonstrate that CASA accurately identifies insufficient arguments. We further deploy CASA in a writing assistance application, and find that suggestions generated by CASA enhance the sufficiency of student-written arguments. Code and data are available at https://github.com/xxxiaol/CASA.

2023

1. EACL
   2023

   UnifEE: Unified Evidence Extraction for Fact Verification

   Nan Hu, Zirui Wu, Yuxuan Lai, and 2 more authors

   In Proceedings of the 17th Conference of the European Chapter of the Association for Computational Linguistics, May 2023

   Abs

   FEVEROUS is a fact extraction and verification task that requires systems to extract evidence of both sentences and table cells from a Wikipedia dump, then predict the veracity of the given claim accordingly. Existing works extract evidence in the two formats separately, ignoring potential connections between them. In this paper, we propose a Unified Evidence Extraction model (UnifEE), which uses a mixed evidence graph to extract the evidence in both formats. With the carefully-designed unified evidence graph, UnifEE allows evidence interactions among all candidates in both formats at similar granularity. Experiments show that, with information aggregated from related evidence candidates in the fusion graph, UnifEE can make better decisions about which evidence should be kept, especially for claims requiring multi-hop reasoning or a combination of tables and texts. Thus it outperforms all previous evidence extraction methods and brings significant improvement in the subsequent claim verification step.

2. EACL
   2023

   An Intra-Class Relation Guided Approach for Code Comment Generation

   Zhenni Wang, Xiaohan Yu, Yansong Feng, and 1 more author

   In Findings of the Association for Computational Linguistics: EACL 2023, May 2023

   Abs

   Code comments are critical for maintaining and comprehending software programs, but they are often missing, mismatched, or outdated in practice. Code comment generation task aims to automatically produce descriptive comments for code snippets. Recently, methods based on the neural encoder-decoder architecture have achieved impressive performance. These methods assume that all the information required to generate comments is encoded in the target function itself, yet in most realistic situations, it is hard to understand a function in isolation from the surrounding context. Furthermore, the global context may contain redundant information that should not be introduced. To address the above issues, we present a novel graph-based learning framework to capture various relations among functions in a class file. Our approach is based on a common real-world scenario in which only a few functions in the source file have human-written comments. Guided by intra-class function relations, our model incorporates contextual information extracted from both the source code and available comments to generate missing comments. We conduct experiments on a Java dataset collected from real-world projects. Experimental results show that the proposed method outperforms competitive baseline models on all automatic and human evaluation metrics.

3. EACL
   2023

   Cross-Lingual Question Answering over Knowledge Base as Reading Comprehension

   Chen Zhang, Yuxuan Lai, Yansong Feng, and 3 more authors

   In Findings of the Association for Computational Linguistics: EACL 2023, May 2023

   Abs

   Although many large-scale knowledge bases (KBs) claim to contain multilingual information, their support for many non-English languages is often incomplete. This incompleteness gives birth to the task of cross-lingual question answering over knowledge base (xKBQA), which aims to answer questions in languages different from that of the provided KB. One of the major challenges facing xKBQA is the high cost of data annotation, leading to limited resources available for further exploration. Another challenge is mapping KB schemas and natural language expressions in the questions under cross-lingual settings. In this paper, we propose a novel approach for xKBQA in a reading comprehension paradigm. We convert KB subgraphs into passages to narrow the gap between KB schemas and questions, which enables our model to benefit from recent advances in multilingual pre-trained language models (MPLMs) and cross-lingual machine reading comprehension (xMRC). Specifically, we use MPLMs, with considerable knowledge of cross-lingual mappings, for cross-lingual reading comprehension. Existing high-quality xMRC datasets can be further utilized to finetune our model, greatly alleviating the data scarcity issue in xKBQA. Extensive experiments on two xKBQA datasets in 12 languages show that our approach outperforms various baselines and achieves strong few-shot and zero-shot performance. Our dataset and code are released for further research.

4. ACL
   2023

   Align-then-Enhance: Multilingual Entailment Graph Enhancement with Soft Predicate Alignment

   Yuting Wu, Yutong Hu, Yansong Feng, and 3 more authors

   In Findings of the Association for Computational Linguistics: ACL 2023, Jul 2023

   Abs

   Entailment graphs (EGs) with predicates as nodes and entailment relations as edges are typically incomplete, while EGs in different languages are often complementary to each other. In this paper, we propose a new task, multilingual entailment graph enhancement, which aims to utilize the entailment information from one EG to enhance another EG in a different language. The ultimate goal is to obtain an enhanced EG containing richer and more accurate entailment information. We present an align-then-enhance framework (ATE) to achieve accurate multilingual entailment graph enhancement, which first exploits a cross-graph guided interaction mechanism to automatically discover potential equivalent predicates between different EGs and then constructs more accurate enhanced entailment graphs based on soft predicate alignments. Extensive experiments show that ATE achieves better and more robust predicate alignment results between different EGs, and the enhanced entailment graphs generated by ATE outperform the original graphs for entailment detection.

5. How Many Answers Should I Give? An Empirical Study of Multi-Answer Reading Comprehension

   Chen Zhang, Jiuheng Lin, Xiao Liu, and 3 more authors

   In Findings of the Association for Computational Linguistics: ACL 2023, Jul 2023

   Abs

   The multi-answer phenomenon, where a question may have multiple answers scattered in the document, can be well handled by humans but is challenging enough for machine reading comprehension (MRC) systems. Despite recent progress in multi-answer MRC, there lacks a systematic analysis of how this phenomenon arises and how to better address it. In this work, we design a taxonomy to categorize commonly-seen multi-answer MRC instances, with which we inspect three multi-answer datasets and analyze where the multi-answer challenge comes from. We further analyze how well different paradigms of current multi-answer MRC models deal with different types of multi-answer instances. We find that some paradigms capture well the key information in the questions while others better model the relation between questions and contexts. We thus explore strategies to make the best of the strengths of different paradigms. Experiments show that generation models can be a promising platform to incorporate different paradigms. Our annotations and code are released for further research.

6. Structure-Discourse Hierarchical Graph for Conditional Question Answering on Long Documents

   Haowei Du, Yansong Feng, Chen Li, and 3 more authors

   In Findings of the Association for Computational Linguistics: ACL 2023, Jul 2023

   Abs

   Conditional question answering on long documents aims to find probable answers and identify conditions that need to be satisfied to make the answers correct over long documents. Existing approaches solve this task by segmenting long documents into multiple sections, and attending information at global and local tokens to predict the answers and corresponding conditions. However, the natural structure of the document and discourse relations between sentences in each document section are ignored, which are crucial for condition retrieving across sections, as well as logical interaction over the question and conditions. To address this issue, this paper constructs a Structure-Discourse Hierarchical Graph (SDHG) and conducts bottom-up information propagation. Firstly we build the sentence-level discourse graphs for each section and encode the discourse relations by graph attention. Secondly, we construct a section-level structure graph based on natural structures, and conduct interactions over the question and contexts. Finally different levels of representations are integrated into jointly answer and condition decoding. The experiments on the benchmark ConditionalQA shows our approach gains over the prior state-of-the-art, by 3.0 EM score and 2.4 F1 score on answer measuring, as well as 2.2 EM score and 1.9 F1 score on jointly answer and condition measuring.

7. The Magic of IF: Investigating Causal Reasoning Abilities in Large Language Models of Code

   Xiao Liu, Da Yin, Chen Zhang, and 2 more authors

   In Findings of the Association for Computational Linguistics: ACL 2023, Jul 2023

   Abs

   Causal reasoning, the ability to identify cause-and-effect relationship, is crucial in human thinking. Although large language models (LLMs) succeed in many NLP tasks, it is still challenging for them to conduct complex causal reasoning like abductive reasoning and counterfactual reasoning. Given the fact that programming code may express causal relations more often and explicitly with conditional statements like “if“, we want to explore whether Code-LLMs acquire better causal reasoning abilities. Our experiments show that compared to text-only LLMs, Code-LLMs with code prompts are better causal reasoners. We further intervene on the prompts from different aspects, and discover that the key point is the programming structure. Code and data are available at \urlhttps://github.com/xxxiaol/magic-if.

8. Learning Dynamic Representations for Discourse Dependency Parsing

   Tianyi Liu, Yansong Feng, and Dongyan Zhao

   In Findings of the Association for Computational Linguistics: EMNLP 2023, Dec 2023

   Abs

   Transition systems have been widely used for the discourse dependency parsing task. Existing works often characterize transition states by examining a certain number of elementary discourse units (EDUs), while neglecting the arcs obtained from the transition history. In this paper, we propose to employ GAT-based encoder to learn dynamic representations for sub-trees constructed in previous transition steps. By incorporating these representations, our model is able to retain accessibility to all parsed EDUs through the obtained arcs, thus better utilizing the structural information of the document, particularly when handling lengthy text spans with complex structures. For the discourse relation recognition task, we employ edge-featured GATs to derive better representations for EDU pairs. Experimental results show that our model can achieve state-of-the-art performance on widely adopted datasets including RST-DT, SciDTB and CDTB. Our code is available at https://github.com/lty-lty/Discourse-Dependency-Parsing.

9. From the One, Judge of the Whole: Typed Entailment Graph Construction with Predicate Generation

   Zhibin Chen, Yansong Feng, and Dongyan Zhao

   In Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), Jul 2023

   Abs

   Entailment Graphs (EGs) have been constructed based on extracted corpora as a strong and explainable form to indicate context-independent entailment relation in natural languages. However, EGs built by previous methods often suffer from the severe sparsity issues, due to limited corpora available and the long-tail phenomenon of predicate distributions. In this paper, we propose a multi-stage method, Typed Predicate-Entailment Graph Generator (TP-EGG), to tackle this problem. Given several seed predicates, TP-EGG builds the graphs by generating new predicates and detecting entailment relations among them. The generative nature of TP-EGG helps us leverage the recent advances from large pretrained language models (PLMs), while avoiding the reliance on carefully prepared corpora. Experiments on benchmark datasets show that TP-EGG can generate high-quality and scale-controllable entailment graphs, achieving significant in-domain improvement over state-of-the-art EGs and boosting the performance of down-stream inference tasks.

10. More than Classification: A Unified Framework for Event Temporal Relation Extraction

    Quzhe Huang, Yutong Hu, Shengqi Zhu, and 3 more authors

    In Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), Jul 2023

    Abs

    Event temporal relation extraction (ETRE) is usually formulated as a multi-label classification task, where each type of relation is simply treated as a one-hot label. This formulation ignores the meaning of relations and wipes out their intrinsic dependency. After examining the relation definitions in various ETRE tasks, we observe that all relations can be interpreted using the start and end time points of events. For example, relation \textitIncludes could be interpreted as event 1 starting no later than event 2 and ending no earlier than event 2. In this paper, we propose a unified event temporal relation extraction framework, which transforms temporal relations into logical expressions of time points and completes the ETRE by predicting the relations between certain time point pairs. Experiments on TB-Dense and MATRES show significant improvements over a strong baseline and outperform the state-of-the-art model by 0.3% on both datasets. By representing all relations in a unified framework, we can leverage the relations with sufficient data to assist the learning of other relations, thus achieving stable improvement in low-data scenarios. When the relation definitions are changed, our method can quickly adapt to the new ones by simply modifying the logic expressions that map time points to new event relations. The code is released at \urlhttps://github.com/AndrewZhe/A-Unified-Framework-for-ETRE

11. Enhancing Structured Evidence Extraction for Fact Verification

    Zirui Wu, Nan Hu, and Yansong Feng

    In Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing, Dec 2023

    Abs

    Open-domain fact verification is the task of verifying claims in natural language texts against extracted evidence. FEVEROUS is a benchmark that requires extracting and integrating both unstructured and structured evidence to verify a given claim. Previous models suffer from low recall of structured evidence extraction, i.e., table extraction and cell selection. In this paper, we propose a simple but effective method to enhance the extraction of structured evidence by leveraging the row and column semantics of tables. Our method comprises two components: (i) a coarse-grained table extraction module that selects tables based on rows and columns relevant to the claim and (ii) a fine-grained cell selection graph that combines both formats of evidence and enables multi-hop and numerical reasoning. We evaluate our method on FEVEROUS and achieve an evidence recall of 60.01% on the test set, which is 6.14% higher than the previous state-of-the-art performance. Our results demonstrate that our method can extract tables and select cells effectively, and provide better evidence sets for verdict prediction. Our code is released at https://github.com/WilliamZR/see-st

12. DiNeR: A Large Realistic Dataset for Evaluating Compositional Generalization

    Chengang Hu, Xiao Liu, and Yansong Feng

    In Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing, Dec 2023

    Abs

    Most of the existing compositional generalization datasets are synthetically-generated, resulting in a lack of natural language variation. While there have been recent attempts to introduce non-synthetic datasets for compositional generalization, they suffer from either limited data scale or a lack of diversity in the forms of combinations. To better investigate compositional generalization with more linguistic phenomena and compositional diversity, we propose the DIsh NamE Recognition (DiNeR) task and create a large realistic Chinese dataset. Given a recipe instruction, models are required to recognize the dish name composed of diverse combinations of food, actions, and flavors. Our dataset consists of 3,811 dishes and 228,114 recipes, and involves plenty of linguistic phenomena such as anaphora, omission and ambiguity. We provide two strong baselines based on T5 and large language models (LLMs). This work contributes a challenging task, baseline methods to tackle the task, and insights into compositional generalization in the context of dish name recognition.

13. Can bert refrain from forgetting on sequential tasks? a probing study

    Mingxu Tao, Yansong Feng, and Dongyan Zhao

    In The Eleventh International Conference on Learning Representations, Dec 2023
14. A frustratingly easy improvement for position embeddings via random padding

    Mingxu Tao, Yansong Feng, and Dongyan Zhao

    In CCF International Conference on Natural Language Processing and Chinese Computing, Dec 2023
15. Lawyer llama technical report

    Quzhe Huang, Mingxu Tao, Chen Zhang, and 5 more authors

    arXiv preprint arXiv:2305.15062, Dec 2023

2022

1. Things not Written in Text: Exploring Spatial Commonsense from Visual Signals

   Xiao Liu, Da Yin, Yansong Feng, and 1 more author

   In Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), May 2022

   Abs

   Spatial commonsense, the knowledge about spatial position and relationship between objects (like the relative size of a lion and a girl, and the position of a boy relative to a bicycle when cycling), is an important part of commonsense knowledge. Although pretrained language models (PLMs) succeed in many NLP tasks, they are shown to be ineffective in spatial commonsense reasoning. Starting from the observation that images are more likely to exhibit spatial commonsense than texts, we explore whether models with visual signals learn more spatial commonsense than text-based PLMs. We propose a spatial commonsense benchmark that focuses on the relative scales of objects, and the positional relationship between people and objects under different actions. We probe PLMs and models with visual signals, including vision-language pretrained models and image synthesis models, on this benchmark, and find that image synthesis models are more capable of learning accurate and consistent spatial knowledge than other models. The spatial knowledge from image synthesis models also helps in natural language understanding tasks that require spatial commonsense.

2. Entailment Graph Learning with Textual Entailment and Soft Transitivity

   Zhibin Chen, Yansong Feng, and Dongyan Zhao

   In Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), May 2022

   Abs

   Typed entailment graphs try to learn the entailment relations between predicates from text and model them as edges between predicate nodes. The construction of entailment graphs usually suffers from severe sparsity and unreliability of distributional similarity. We propose a two-stage method, Entailment Graph with Textual Entailment and Transitivity (EGT2). EGT2 learns the local entailment relations by recognizing the textual entailment between template sentences formed by typed CCG-parsed predicates. Based on the generated local graph, EGT2 then uses three novel soft transitivity constraints to consider the logical transitivity in entailment structures. Experiments on benchmark datasets show that EGT2 can well model the transitivity in entailment graph to alleviate the sparsity, and leads to signifcant improvement over current state-of-the-art methods.

3. Does Recommend-Revise Produce Reliable Annotations? An Analysis on Missing Instances in DocRED

   Quzhe Huang, Shibo Hao, Yuan Ye, and 3 more authors

   In Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), May 2022

   Abs

   DocRED is a widely used dataset for document-level relation extraction. In the large-scale annotation, a recommend-revise scheme is adopted to reduce the workload. Within this scheme, annotators are provided with candidate relation instances from distant supervision, and they then manually supplement and remove relational facts based on the recommendations. However, when comparing DocRED with a subset relabeled from scratch, we find that this scheme results in a considerable amount of false negative samples and an obvious bias towards popular entities and relations. Furthermore, we observe that the models trained on DocRED have low recall on our relabeled dataset and inherit the same bias in the training data. Through the analysis of annotators’ behaviors, we figure out the underlying reason for the problems above: the scheme actually discourages annotators from supplementing adequate instances in the revision phase. We appeal to future research to take into consideration the issues with the recommend-revise scheme when designing new models and annotation schemes. The relabeled dataset is released at \urlhttps://github.com/AndrewZhe/Revisit-DocRED, to serve as a more reliable test set of document RE models.

4. Improve Discourse Dependency Parsing with Contextualized Representations

   Yifei Zhou, and Yansong Feng

   In Findings of the Association for Computational Linguistics: NAACL 2022, Jul 2022

   Abs

   Previous works show that discourse analysis benefits from modeling intra- and inter-sentential levels separately, where proper representations for text units of different granularities are desired to capture both the information of the text units and their relation to the context. In this paper, we propose to take advantage of transformers to encode different contextualized representations of units of different levels to dynamically capture the information required for discourse dependency analysis on intra- and inter-sentential levels. Motivated by the observation of writing patterns shared across articles to improve discourse analysis, we propose to design sequence labeling methods to take advantage of such structural information from the context that substantially outperforms traditional direct classification methods. Experiments show that our model achieves state-of-the-art results on both English and Chinese datasets.

5. Do Charge Prediction Models Learn Legal Theory?

   Zhenwei An, Quzhe Huang, Cong Jiang, and 2 more authors

   In Findings of the Association for Computational Linguistics: EMNLP 2022, Dec 2022

   Abs

   The charge prediction task aims to predict the charge for a case given its fact description. Recent models have already achieved impressive accuracy in this task, however, little is understood about the mechanisms they use to perform the judgment.For practical applications, a charge prediction model should conform to the certain legal theory in civil law countries, as under the framework of civil law, all cases are judged according to certain local legal theories. In China, for example, nearly all criminal judges make decisions based on the Four Elements Theory (FET).In this paper, we argue that trustworthy charge prediction models should take legal theories into consideration, and standing on prior studies in model interpretation, we propose three principles for trustworthy models should follow in this task, which are sensitive, selective, and presumption of innocence.We further design a new framework to evaluate whether existing charge prediction models learn legal theories. Our findings indicate that, while existing charge prediction models meet the selective principle on a benchmark dataset, most of them are still not sensitive enough and do not satisfy the presumption of innocence. Our code and dataset are released at \urlhttps://github.com/ZhenweiAn/EXP_LJP.

6. Dual-Channel Evidence Fusion for Fact Verification over Texts and Tables

   Nan Hu, Zirui Wu, Yuxuan Lai, and 2 more authors

   In Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Jul 2022

   Abs

   Different from previous fact extraction and verification tasks that only consider evidence of a single format, FEVEROUS brings further challenges by extending the evidence format to both plain text and tables. Existing works convert all candidate evidence into either sentences or tables, thus often failing to fully capture the rich context in their original format from the converted evidence, let alone the context information lost during conversion. In this paper, we propose a Dual Channel Unified Format fact verification model (DCUF), which unifies various evidence into parallel streams, i.e., natural language sentences and a global evidence table, simultaneously. With carefully-designed evidence conversion and organization methods, DCUF makes the most of pre-trained table/language models to encourage each evidence piece to perform early and thorough interactions with other pieces in its original format. Experiments show that our model can make better use of existing pre-trained models to absorb evidence of two formats, thus outperforming previous works by a large margin. Our code and models are publicly available.

7. Counterfactual Recipe Generation: Exploring Compositional Generalization in a Realistic Scenario

   Xiao Liu, Yansong Feng, Jizhi Tang, and 2 more authors

   In Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing, Dec 2022

   Abs

   People can acquire knowledge in an unsupervised manner by reading, and compose the knowledge to make novel combinations. In this paper, we investigate whether pretrained language models can perform compositional generalization in a realistic setting: recipe generation. We design the counterfactual recipe generation task, which asks models to modify a base recipe according to the change of an ingredient. This task requires compositional generalization at two levels: the surface level of incorporating the new ingredient into the base recipe, and the deeper level of adjusting actions related to the changing ingredient. We collect a large-scale recipe dataset in Chinese for models to learn culinary knowledge, and a subset of action-level fine-grained annotations for evaluation.We finetune pretrained language models on the recipe corpus, and use unsupervised counterfactual generation methods to generate modified recipes.Results show that existing models have difficulties in modifying the ingredients while preserving the original text style, and often miss actions that need to be adjusted. Although pretrained language models can generate fluent recipe texts, they fail to truly learn and use the culinary knowledge in a compositional way. Code and data are available at https://github.com/xxxiaol/counterfactual-recipe-generation.

8. Integrating manifold knowledge for global entity linking with heterogeneous graphs

   Zhibin Chen, Yuting Wu, Yansong Feng, and 1 more author

   Data Intelligence, Dec 2022

2021

1. Why Machine Reading Comprehension Models Learn Shortcuts?

   Yuxuan Lai, Chen Zhang, Yansong Feng, and 2 more authors

   In Findings of the Association for Computational Linguistics: ACL-IJCNLP 2021, Aug 2021
2. Extract, Integrate, Compete: Towards Verification Style Reading Comprehension

   Chen Zhang, Yuxuan Lai, Yansong Feng, and 1 more author

   In Findings of the Association for Computational Linguistics: EMNLP 2021, Nov 2021

   Abs

   In this paper, we present a new verification style reading comprehension dataset named VGaokao from Chinese Language tests of Gaokao. Different from existing efforts, the new dataset is originally designed for native speakers’ evaluation, thus requiring more advanced language understanding skills. To address the challenges in VGaokao, we propose a novel Extract-Integrate-Compete approach, which iteratively selects complementary evidence with a novel query updating mechanism and adaptively distills supportive evidence, followed by a pairwise competition to push models to learn the subtle difference among similar text pieces. Experiments show that our methods outperform various baselines on VGaokao with retrieved complementary evidence, while having the merits of efficiency and explainability. Our dataset and code are released for further research.

3. Learning to Organize a Bag of Words into Sentences with Neural Networks: An Empirical Study

   Chongyang Tao, Shen Gao, Juntao Li, and 3 more authors

   In Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Jun 2021

   Abs

   Sequential information, a.k.a., orders, is assumed to be essential for processing a sequence with recurrent neural network or convolutional neural network based encoders. However, is it possible to encode natural languages without orders? Given a bag of words from a disordered sentence, humans may still be able to understand what those words mean by reordering or reconstructing them. Inspired by such an intuition, in this paper, we perform a study to investigate how “order” information takes effects in natural language learning. By running comprehensive comparisons, we quantitatively compare the ability of several representative neural models to organize sentences from a bag of words under three typical scenarios, and summarize some empirical findings and challenges, which can shed light on future research on this line of work.

4. Lattice-BERT: Leveraging Multi-Granularity Representations in Chinese Pre-trained Language Models

   Yuxuan Lai, Yijia Liu, Yansong Feng, and 2 more authors

   In Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Jun 2021

   Abs

   Chinese pre-trained language models usually process text as a sequence of characters, while ignoring more coarse granularity, e.g., words. In this work, we propose a novel pre-training paradigm for Chinese — Lattice-BERT, which explicitly incorporates word representations along with characters, thus can model a sentence in a multi-granularity manner. Specifically, we construct a lattice graph from the characters and words in a sentence and feed all these text units into transformers. We design a lattice position attention mechanism to exploit the lattice structures in self-attention layers. We further propose a masked segment prediction task to push the model to learn from rich but redundant information inherent in lattices, while avoiding learning unexpected tricks. Experiments on 11 Chinese natural language understanding tasks show that our model can bring an average increase of 1.5% under the 12-layer setting, which achieves new state-of-the-art among base-size models on the CLUE benchmarks. Further analysis shows that Lattice-BERT can harness the lattice structures, and the improvement comes from the exploration of redundant information and multi-granularity representations. Our code will be available at \urlhttps://github.com/alibaba/pretrained-language-models/LatticeBERT.

5. Everything Has a Cause: Leveraging Causal Inference in Legal Text Analysis

   Xiao Liu, Da Yin, Yansong Feng, and 2 more authors

   In Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Jun 2021

   Abs

   Causal inference is the process of capturing cause-effect relationship among variables. Most existing works focus on dealing with structured data, while mining causal relationship among factors from unstructured data, like text, has been less examined, but is of great importance, especially in the legal domain. In this paper, we propose a novel Graph-based Causal Inference (GCI) framework, which builds causal graphs from fact descriptions without much human involvement and enables causal inference to facilitate legal practitioners to make proper decisions. We evaluate the framework on a challenging similar charge disambiguation task. Experimental results show that GCI can capture the nuance from fact descriptions among multiple confusing charges and provide explainable discrimination, especially in few-shot settings. We also observe that the causal knowledge contained in GCI can be effectively injected into powerful neural networks for better performance and interpretability.

6. Exploring Distantly-Labeled Rationales in Neural Network Models

   Quzhe Huang, Shengqi Zhu, Yansong Feng, and 1 more author

   In 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), Aug 2021

   Abs

   Recent studies strive to incorporate various human rationales into neural networks to improve model performance, but few pay attention to the quality of the rationales. Most existing methods distribute their models’ focus to distantly-labeled rationale words entirely and equally, while ignoring the potential important non-rationale words and not distinguishing the importance of different rationale words. In this paper, we propose two novel auxiliary loss functions to make better use of distantly-labeled rationales, which encourage models to maintain their focus on important words beyond labeled rationales (PINs) and alleviate redundant training on non-helpful rationales (NoIRs). Experiments on two representative classification tasks show that our proposed methods can push a classification model to effectively learn crucial clues from non-perfect rationales while maintaining the ability to spread its focus to other unlabeled important words, thus significantly outperform existing methods.

7. Three Sentences Are All You Need: Local Path Enhanced Document Relation Extraction

   Quzhe Huang, Shengqi Zhu, Yansong Feng, and 3 more authors

   In Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 2: Short Papers), Aug 2021

   Abs

   Document-level Relation Extraction (RE) is a more challenging task than sentence RE as it often requires reasoning over multiple sentences. Yet, human annotators usually use a small number of sentences to identify the relationship between a given entity pair. In this paper, we present an embarrassingly simple but effective method to heuristically select evidence sentences for document-level RE, which can be easily combined with BiLSTM to achieve good performance on benchmark datasets, even better than fancy graph neural network based methods. We have released our code at \urlhttps://github.com/AndrewZhe/Three-Sentences-Are-All-You-Need.

8. A review of deep learning in question answering over knowledge bases

   Chen Zhang, Yuxuan Lai, Yansong Feng, and 1 more author

   AI Open, Aug 2021

2020

1. Semantic Graphs for Generating Deep Questions

   Liangming Pan, Yuxi Xie, Yansong Feng, and 2 more authors

   In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, Jul 2020

   Abs

   This paper proposes the problem of Deep Question Generation (DQG), which aims to generate complex questions that require reasoning over multiple pieces of information about the input passage. In order to capture the global structure of the document and facilitate reasoning, we propose a novel framework that first constructs a semantic-level graph for the input document and then encodes the semantic graph by introducing an attention-based GGNN (Att-GGNN). Afterward, we fuse the document-level and graph-level representations to perform joint training of content selection and question decoding. On the HotpotQA deep-question centric dataset, our model greatly improves performance over questions requiring reasoning over multiple facts, leading to state-of-the-art performance. The code is publicly available at \urlhttps://github.com/WING-NUS/SG-Deep-Question-Generation.

2. Neighborhood Matching Network for Entity Alignment

   Yuting Wu, Xiao Liu, Yansong Feng, and 2 more authors

   In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, Jul 2020

   Abs

   Structural heterogeneity between knowledge graphs is an outstanding challenge for entity alignment. This paper presents Neighborhood Matching Network (NMN), a novel entity alignment framework for tackling the structural heterogeneity challenge. NMN estimates the similarities between entities to capture both the topological structure and the neighborhood difference. It provides two innovative components for better learning representations for entity alignment. It first uses a novel graph sampling method to distill a discriminative neighborhood for each entity. It then adopts a cross-graph neighborhood matching module to jointly encode the neighborhood difference for a given entity pair. Such strategies allow NMN to effectively construct matching-oriented entity representations while ignoring noisy neighbors that have a negative impact on the alignment task. Extensive experiments performed on three entity alignment datasets show that NMN can well estimate the neighborhood similarity in more tough cases and significantly outperforms 12 previous state-of-the-art methods.

3. Exploring Question-Specific Rewards for Generating Deep Questions

   Yuxi Xie, Liangming Pan, Dongzhe Wang, and 2 more authors

   In Proceedings of the 28th International Conference on Computational Linguistics, Dec 2020

   Abs

   Recent question generation (QG) approaches often utilize the sequence-to-sequence framework (Seq2Seq) to optimize the log likelihood of ground-truth questions using teacher forcing. However, this training objective is inconsistent with actual question quality, which is often reflected by certain global properties such as whether the question can be answered by the document. As such, we directly optimize for QG-specific objectives via reinforcement learning to improve question quality. We design three different rewards that target to improve the fluency, relevance, and answerability of generated questions. We conduct both automatic and human evaluations in addition to thorough analysis to explore the effect of each QG-specific reward. We find that optimizing on question-specific rewards generally leads to better performance in automatic evaluation metrics. However, only the rewards that correlate well with human judgement (e.g., relevance) lead to real improvement in question quality. Optimizing for the others, especially answerability, introduces incorrect bias to the model, resulting in poorer question quality. The code is publicly available at \urlhttps://github.com/YuxiXie/RL-for-Question-Generation.

4. Understanding Procedural Text using Interactive Entity Networks

   Jizhi Tang, Yansong Feng, and Dongyan Zhao

   In Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP), Nov 2020

   Abs

   The task of procedural text comprehension aims to understand the dynamic nature of entities/objects in a process. Here, the key is to track how the entities interact with each other and how their states are changing along the procedure. Recent efforts have made great progress to track multiple entities in a procedural text, but usually treat each entity separately and ignore the fact that there are often multiple entities interacting with each other during one process, some of which are even explicitly mentioned. In this paper, we propose a novel Interactive Entity Network (IEN), which is a recurrent network with memory equipped cells for state tracking. In each IEN cell, we maintain different attention matrices through specific memories to model different types of entity interactions. Importantly, we can update these memories in a sequential manner so as to explore the causal relationship between entity actions and subsequent state changes. We evaluate our model on a benchmark dataset, and the results show that IEN outperforms state-of-the-art models by precisely capturing the interactions of multiple entities and explicitly leverage the relationship between entity interactions and subsequent state changes.

5. Towards Context-Aware Code Comment Generation

   Xiaohan Yu, Quzhe Huang, Zheng Wang, and 2 more authors

   In Findings of the Association for Computational Linguistics: EMNLP 2020, Nov 2020

   Abs

   Code comments are vital for software maintenance and comprehension, but many software projects suffer from the lack of meaningful and up-to-date comments in practice. This paper presents a novel approach to automatically generate code comments at a function level by targeting object-oriented programming languages. Unlike prior work that only uses information locally available within the target function, our approach leverages broader contextual information by considering all other functions of the same class. To propagate and integrate information beyond the scope of the target function, we design a novel learning framework based on the bidirectional gated recurrent unit and a graph attention network with a pointer mechanism. We apply our approach to produce code comments for Java methods and compare it against four strong baseline methods. Experimental results show that our approach outperforms most methods by a large margin and achieves a comparable result with the state-of-the-art method.

6. Combining graph-based learning with automated data collection for code vulnerability detection

   Huanting Wang, Guixin Ye, Zhanyong Tang, and 6 more authors

   IEEE Transactions on Information Forensics and Security, Nov 2020
7. Integrating relation constraints with neural relation extractors

   Yuan Ye, Yansong Feng, Bingfeng Luo, and 2 more authors

   In Proceedings of the AAAI Conference on Artificial Intelligence, Nov 2020
8. Improving matching models with hierarchical contextualized representations for multi-turn response selection

   Chongyang Tao, Wei Wu, Yansong Feng, and 2 more authors

   In Proceedings of the 43rd International ACM SIGIR Conference on Research and Development in Information Retrieval, Nov 2020
9. Coordinated reasoning for cross-lingual knowledge graph alignment

   Kun Xu, Linfeng Song, Yansong Feng, and 2 more authors

   In Proceedings of the AAAI conference on artificial intelligence, Nov 2020
10. Using generative adversarial networks to break and protect text captchas

    Guixin Ye, Zhanyong Tang, Dingyi Fang, and 6 more authors

    ACM Transactions on Privacy and Security (TOPS), Nov 2020
11. Simplifying Graph Attention Networks with Source-Target Separation

    Hantao Guo, Rui Yan, Yansong Feng, and 2 more authors

    In ECAI 2020, Nov 2020
12. Domain adaptation for semantic parsing

    Zechang Li, Yuxuan Lai, Yansong Feng, and 1 more author

    arXiv preprint arXiv:2006.13071, Nov 2020
13. Multi-turn intent determination and slot filling with neural networks and regular expressions

    Waheed Ahmed Abro, Guilin Qi, Zafar Ali, and 2 more authors

    Knowledge-Based Systems, Nov 2020
14. Latent template induction with Gumbel-CRFS

    Yao Fu, Chuanqi Tan, Bin Bi, and 3 more authors

    Advances in Neural Information Processing Systems, Nov 2020

2019

1. Jointly Learning Entity and Relation Representations for Entity Alignment

   Yuting Wu, Xiao Liu, Yansong Feng, and 2 more authors

   In Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP), Nov 2019

   Abs

   Entity alignment is a viable means for integrating heterogeneous knowledge among different knowledge graphs (KGs). Recent developments in the field often take an embedding-based approach to model the structural information of KGs so that entity alignment can be easily performed in the embedding space. However, most existing works do not explicitly utilize useful relation representations to assist in entity alignment, which, as we will show in the paper, is a simple yet effective way for improving entity alignment. This paper presents a novel joint learning framework for entity alignment. At the core of our approach is a Graph Convolutional Network (GCN) based framework for learning both entity and relation representations. Rather than relying on pre-aligned relation seeds to learn relation representations, we first approximate them using entity embeddings learned by the GCN. We then incorporate the relation approximation into entities to iteratively learn better representations for both. Experiments performed on three real-world cross-lingual datasets show that our approach substantially outperforms state-of-the-art entity alignment methods.

2. Sampling Matters! An Empirical Study of Negative Sampling Strategies for Learning of Matching Models in Retrieval-based Dialogue Systems

   Jia Li, Chongyang Tao, Wei Wu, and 3 more authors

   In Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP), Nov 2019

   Abs

   We study how to sample negative examples to automatically construct a training set for effective model learning in retrieval-based dialogue systems. Following an idea of dynamically adapting negative examples to matching models in learning, we consider four strategies including minimum sampling, maximum sampling, semi-hard sampling, and decay-hard sampling. Empirical studies on two benchmarks with three matching models indicate that compared with the widely used random sampling strategy, although the first two strategies lead to performance drop, the latter two ones can bring consistent improvement to the performance of all the models on both benchmarks.

3. Learning to Update Knowledge Graphs by Reading News

   Jizhi Tang, Yansong Feng, and Dongyan Zhao

   In Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP), Nov 2019

   Abs

   News streams contain rich up-to-date information which can be used to update knowledge graphs (KGs). Most current text-based KG updating methods rely on elaborately designed information extraction (IE) systems and carefully crafted rules, which are often domain-specific and hard to maintain. Besides, such methods often hardly pay enough attention to the implicit information that lies underneath texts. In this paper, we propose a novel neural network method, GUpdater, to tackle these problems. GUpdater is built upon graph neural networks (GNNs) with a text-based attention mechanism to guide the updating message passing through the KG structures. Experiments on a real-world KG updating dataset show that our model can effectively broadcast the news information to the KG structures and perform necessary link-adding or link-deleting operations to ensure the KG up-to-date according to news snippets.

4. Easy First Relation Extraction with Information Redundancy

   Shuai Ma, Gang Wang, Yansong Feng, and 1 more author

   In Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP), Nov 2019

   Abs

   Many existing relation extraction (RE) models make decisions globally using integer linear programming (ILP). However, it is nontrivial to make use of integer linear programming as a blackbox solver for RE. Its cost of time and memory may become unacceptable with the increase of data scale, and redundant information needs to be encoded cautiously for ILP. In this paper, we propose an easy first approach for relation extraction with information redundancies, embedded in the results produced by local sentence level extractors, during which conflict decisions are resolved with domain and uniqueness constraints. Information redundancies are leveraged to support both easy first collective inference for easy decisions in the first stage and ILP for hard decisions in a subsequent stage. Experimental study shows that our approach improves the efficiency and accuracy of RE, and outperforms both ILP and neural network-based methods.

5. Generating Classical Chinese Poems from Vernacular Chinese

   Zhichao Yang, Pengshan Cai, Yansong Feng, and 4 more authors

   In Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP), Nov 2019

   Abs

   Classical Chinese poetry is a jewel in the treasure house of Chinese culture. Previous poem generation models only allow users to employ keywords to interfere the meaning of generated poems, leaving the dominion of generation to the model. In this paper, we propose a novel task of generating classical Chinese poems from vernacular, which allows users to have more control over the semantic of generated poems. We adapt the approach of unsupervised machine translation (UMT) to our task. We use segmentation-based padding and reinforcement learning to address under-translation and over-translation respectively. According to experiments, our approach significantly improve the perplexity and BLEU compared with typical UMT models. Furthermore, we explored guidelines on how to write the input vernacular to generate better poems. Human evaluation showed our approach can generate high-quality poems which are comparable to amateur poems.

6. Enhancing Key-Value Memory Neural Networks for Knowledge Based Question Answering

   Kun Xu, Yuxuan Lai, Yansong Feng, and 1 more author

   In Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers), Jun 2019

   Abs

   Traditional Key-value Memory Neural Networks (KV-MemNNs) are proved to be effective to support shallow reasoning over a collection of documents in domain specific Question Answering or Reading Comprehension tasks. However, extending KV-MemNNs to Knowledge Based Question Answering (KB-QA) is not trivia, which should properly decompose a complex question into a sequence of queries against the memory, and update the query representations to support multi-hop reasoning over the memory. In this paper, we propose a novel mechanism to enable conventional KV-MemNNs models to perform interpretable reasoning for complex questions. To achieve this, we design a new query updating strategy to mask previously-addressed memory information from the query representations, and introduce a novel STOP strategy to avoid invalid or repeated memory reading without strong annotation signals. This also enables KV-MemNNs to produce structured queries and work in a semantic parsing fashion. Experimental results on benchmark datasets show that our solution, trained with question-answer pairs only, can provide conventional KV-MemNNs models with better reasoning abilities on complex questions, and achieve state-of-art performances.

7. Cross-lingual Knowledge Graph Alignment via Graph Matching Neural Network

   Kun Xu, Liwei Wang, Mo Yu, and 4 more authors

   In Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics, Jul 2019

   Abs

   Previous cross-lingual knowledge graph (KG) alignment studies rely on entity embeddings derived only from monolingual KG structural information, which may fail at matching entities that have different facts in two KGs. In this paper, we introduce the topic entity graph, a local sub-graph of an entity, to represent entities with their contextual information in KG. From this view, the KB-alignment task can be formulated as a graph matching problem; and we further propose a graph-attention based solution, which first matches all entities in two topic entity graphs, and then jointly model the local matching information to derive a graph-level matching vector. Experiments show that our model outperforms previous state-of-the-art methods by a large margin.

8. Learning a Matching Model with Co-teaching for Multi-turn Response Selection in Retrieval-based Dialogue Systems

   Jiazhan Feng, Chongyang Tao, Wei Wu, and 3 more authors

   In Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics, Jul 2019

   Abs

   We study learning of a matching model for response selection in retrieval-based dialogue systems. The problem is equally important with designing the architecture of a model, but is less explored in existing literature. To learn a robust matching model from noisy training data, we propose a general co-teaching framework with three specific teaching strategies that cover both teaching with loss functions and teaching with data curriculum. Under the framework, we simultaneously learn two matching models with independent training sets. In each iteration, one model transfers the knowledge learned from its training set to the other model, and at the same time receives the guide from the other model on how to overcome noise in training. Through being both a teacher and a student, the two models learn from each other and get improved together. Evaluation results on two public data sets indicate that the proposed learning approach can generally and significantly improve the performance of existing matching models.

9. Paraphrase generation with latent bag of words

   Yao Fu, Yansong Feng, and John P Cunningham

   Advances in Neural Information Processing Systems, Jul 2019
10. Lattice cnns for matching based chinese question answering

    Yuxuan Lai, Yansong Feng, Xiaohan Yu, and 3 more authors

    In Proceedings of the AAAI conference on artificial intelligence, Jul 2019

2018

1. Natural Answer Generation with Heterogeneous Memory

   Yao Fu, and Yansong Feng

   In Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long Papers), Jun 2018

   Abs

   Memory augmented encoder-decoder framework has achieved promising progress for natural language generation tasks. Such frameworks enable a decoder to retrieve from a memory during generation. However, less research has been done to take care of the memory contents from different sources, which are often of heterogeneous formats. In this work, we propose a novel attention mechanism to encourage the decoder to actively interact with the memory by taking its heterogeneity into account. Our solution attends across the generated history and memory to explicitly avoid repetition, and introduce related knowledge to enrich our generated sentences. Experiments on the answer sentence generation task show that our method can effectively explore heterogeneous memory to produce readable and meaningful answer sentences while maintaining high coverage for given answer information.

2. SQL-to-Text Generation with Graph-to-Sequence Model

   Kun Xu, Lingfei Wu, Zhiguo Wang, and 2 more authors

   In Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing, Oct 2018

   Abs

   Previous work approaches the SQL-to-text generation task using vanilla Seq2Seq models, which may not fully capture the inherent graph-structured information in SQL query. In this paper, we propose a graph-to-sequence model to encode the global structure information into node embeddings. This model can effectively learn the correlation between the SQL query pattern and its interpretation. Experimental results on the WikiSQL dataset and Stackoverflow dataset show that our model outperforms the Seq2Seq and Tree2Seq baselines, achieving the state-of-the-art performance.

3. Multi-grained Attention Network for Aspect-Level Sentiment Classification

   Feifan Fan, Yansong Feng, and Dongyan Zhao

   In Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing, Oct 2018

   Abs

   We propose a novel multi-grained attention network (MGAN) model for aspect level sentiment classification. Existing approaches mostly adopt coarse-grained attention mechanism, which may bring information loss if the aspect has multiple words or larger context. We propose a fine-grained attention mechanism, which can capture the word-level interaction between aspect and context. And then we leverage the fine-grained and coarse-grained attention mechanisms to compose the MGAN framework. Moreover, unlike previous works which train each aspect with its context separately, we design an aspect alignment loss to depict the aspect-level interactions among the aspects that have the same context. We evaluate the proposed approach on three datasets: laptop and restaurant are from SemEval 2014, and the last one is a twitter dataset. Experimental results show that the multi-grained attention network consistently outperforms the state-of-the-art methods on all three datasets. We also conduct experiments to evaluate the effectiveness of aspect alignment loss, which indicates the aspect-level interactions can bring extra useful information and further improve the performance.

4. Marrying Up Regular Expressions with Neural Networks: A Case Study for Spoken Language Understanding

   Bingfeng Luo, Yansong Feng, Zheng Wang, and 3 more authors

   In Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), Jul 2018

   Abs

   The success of many natural language processing (NLP) tasks is bound by the number and quality of annotated data, but there is often a shortage of such training data. In this paper, we ask the question: “Can we combine a neural network (NN) with regular expressions (RE) to improve supervised learning for NLP?”. In answer, we develop novel methods to exploit the rich expressiveness of REs at different levels within a NN, showing that the combination significantly enhances the learning effectiveness when a small number of training examples are available. We evaluate our approach by applying it to spoken language understanding for intent detection and slot filling. Experimental results show that our approach is highly effective in exploiting the available training data, giving a clear boost to the RE-unaware NN.

5. Modeling discourse cohesion for discourse parsing via memory network

   Yanyan Jia, Yuan Ye, Yansong Feng, and 3 more authors

   In Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers), Jul 2018

   Abs

   Identifying long-span dependencies between discourse units is crucial to improve discourse parsing performance. Most existing approaches design sophisticated features or exploit various off-the-shelf tools, but achieve little success. In this paper, we propose a new transition-based discourse parser that makes use of memory networks to take discourse cohesion into account. The automatically captured discourse cohesion benefits discourse parsing, especially for long span scenarios. Experiments on the RST discourse treebank show that our method outperforms traditional featured based methods, and the memory based discourse cohesion can improve the overall parsing performance significantly.

6. Overview of CAIL2018: Legal judgment prediction competition

   Haoxi Zhong, Chaojun Xiao, Zhipeng Guo, and 8 more authors

   arXiv preprint arXiv:1810.05851, Jul 2018
7. Cail2018: A large-scale legal dataset for judgment prediction

   Chaojun Xiao, Haoxi Zhong, Zhipeng Guo, and 8 more authors

   arXiv preprint arXiv:1807.02478, Jul 2018
8. Graph2seq: Graph to sequence learning with attention-based neural networks

   Kun Xu, Lingfei Wu, Zhiguo Wang, and 3 more authors

   arXiv preprint arXiv:1804.00823, Jul 2018
9. Proteus: Network-aware web browsing on heterogeneous mobile systems

   Jie Ren, Xiaoming Wang, Jianbin Fang, and 5 more authors

   In Proceedings of the 14th International Conference on emerging Networking EXperiments and Technologies, Jul 2018
10. Adaptive web browsing on mobile heterogeneous multi-cores

    Jie Ren, Xiaoming Wang, Jianbin Fang, and 2 more authors

    IEEE Computer Architecture Letters, Jul 2018
11. Scale up event extraction learning via automatic training data generation

    Ying Zeng, Yansong Feng, Rong Ma, and 4 more authors

    In Proceedings of the AAAI Conference on Artificial Intelligence, Jul 2018
12. A chinese question answering system for single-relation factoid questions

    Yuxuan Lai, Yanyan Jia, Yang Lin, and 2 more authors

    In Natural Language Processing and Chinese Computing: 6th CCF International Conference, NLPCC 2017, Dalian, China, November 8–12, 2017, Proceedings 6, Jul 2018
13. Improved discourse parsing with two-step neural transition-based model

    Yanyan Jia, Yansong Feng, Yuan Ye, and 3 more authors

    ACM Transactions on Asian and Low-Resource Language Information Processing (TALLIP), Jul 2018
14. Deep learning in question answering

    Kang Liu, and Yansong Feng

    Deep Learning in Natural Language Processing, Jul 2018
15. Learning to Converse with Noisy Data: Generation with Calibration.

    Mingyue Shang, Zhenxin Fu, Nanyun Peng, and 3 more authors

    In IJCAI, Jul 2018
16. Towards a neural conversation model with diversity net using determinantal point processes

    Yiping Song, Rui Yan, Yansong Feng, and 3 more authors

    In Proceedings of the AAAI Conference on Artificial Intelligence, Jul 2018
17. Yet another text captcha solver: A generative adversarial network based approach

    Guixin Ye, Zhanyong Tang, Dingyi Fang, and 5 more authors

    In Proceedings of the 2018 ACM SIGSAC conference on computer and communications security, Jul 2018
18. Improving matching models with contextualized word representations for multi-turn response selection in retrieval-based chatbots

    Chongyang Tao, Wei Wu, Can Xu, and 3 more authors

    arXiv preprint arXiv:1808.07244, Jul 2018
19. Encoding implicit relation requirements for relation extraction: A joint inference approach

    Liwei Chen, Yansong Feng, Songfang Huang, and 2 more authors

    Artificial Intelligence, Jul 2018

2017

1. Learning with Noise: Enhance Distantly Supervised Relation Extraction with Dynamic Transition Matrix

   Bingfeng Luo, Yansong Feng, Zheng Wang, and 4 more authors

   In Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), Jul 2017
2. How to Make Context More Useful? An Empirical Study on Context-Aware Neural Conversational Models

   Zhiliang Tian, Rui Yan, Lili Mou, and 3 more authors

   In Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers), Jul 2017
3. Towards Implicit Content-Introducing for Generative Short-Text Conversation Systems

   Lili Yao, Yaoyuan Zhang, Yansong Feng, and 2 more authors

   In Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing, Sep 2017
4. Learning to Predict Charges for Criminal Cases with Legal Basis

   Bingfeng Luo, Yansong Feng, Jianbo Xu, and 2 more authors

   In Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing, Sep 2017
5. Learning knowledge representation across knowledge graphs

   Pengshan Cai, Wei Li, Yansong Feng, and 2 more authors

   In Workshops at the Thirty-First AAAI Conference on Artificial Intelligence, Sep 2017

2016

1. Question Answering on Freebase via Relation Extraction and Textual Evidence

   Kun Xu, Siva Reddy, Yansong Feng, and 2 more authors

   In Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), Aug 2016
2. Methods and Theories for Large-scale Structured Prediction

   Xu Sun, and Yansong Feng

   In Proceedings of the 2016 Conference on Empirical Methods in Natural Language Processing: Tutorial Abstracts, Nov 2016
3. Hybrid Question Answering over Knowledge Base and Free Text

   Kun Xu, Yansong Feng, Songfang Huang, and 1 more author

   In Proceedings of COLING 2016, the 26th International Conference on Computational Linguistics: Technical Papers, Dec 2016
4. WIP Event Detection System at TAC KBP 2016 Event Nugget Track.

   Ying Zeng, Bingfeng Luo, Yansong Feng, and 1 more author

   In TAC, Dec 2016
5. Adaptive evolutionary filtering in real-time twitter stream

   Feifan Fan, Yansong Feng, Lili Yao, and 1 more author

   In Proceedings of the 25th ACM International on Conference on Information and Knowledge Management, Dec 2016
6. Detecting Synonymous Predicates from Online Encyclopedia with Rich Features

   Zhe Han, Yansong Feng, and Dongyan Zhao

   In Information Retrieval Technology: 12th Asia Information Retrieval Societies Conference, AIRS 2016, Beijing, China, November 30–December 2, 2016, Proceedings 12, Dec 2016
7. Open domain question answering system based on knowledge base

   Yuxuan Lai, Yang Lin, Jiahao Chen, and 2 more authors

   In Natural Language Understanding and Intelligent Applications: 5th CCF Conference on Natural Language Processing and Chinese Computing, NLPCC 2016, and 24th International Conference on Computer Processing of Oriental Languages, ICCPOL 2016, Kunming, China, December 2–6, 2016, Proceedings 24, Dec 2016
8. Transition-based discourse parsing with multilayer stack long short term memory

   Yanyan Jia, Yansong Feng, Bingfeng Luo, and 3 more authors

   In Natural Language Understanding and Intelligent Applications: 5th CCF Conference on Natural Language Processing and Chinese Computing, NLPCC 2016, and 24th International Conference on Computer Processing of Oriental Languages, ICCPOL 2016, Kunming, China, December 2–6, 2016, Proceedings 24, Dec 2016
9. A convolution BiLSTM neural network model for Chinese event extraction

   Ying Zeng, Honghui Yang, Yansong Feng, and 2 more authors

   In Natural Language Understanding and Intelligent Applications: 5th CCF Conference on Natural Language Processing and Chinese Computing, NLPCC 2016, and 24th International Conference on Computer Processing of Oriental Languages, ICCPOL 2016, Kunming, China, December 2–6, 2016, Proceedings 24, Dec 2016

2015

1. Semantic Relation Classification via Convolutional Neural Networks with Simple Negative Sampling

   Kun Xu, Yansong Feng, Songfang Huang, and 1 more author

   In Proceedings of the 2015 Conference on Empirical Methods in Natural Language Processing, Sep 2015
2. Semantic Interpretation of Superlative Expressions via Structured Knowledge Bases

   Sheng Zhang, Yansong Feng, Songfang Huang, and 3 more authors

   In Proceedings of the 53rd Annual Meeting of the Association for Computational Linguistics and the 7th International Joint Conference on Natural Language Processing (Volume 2: Short Papers), Jul 2015
3. What is the longest river in the usa? semantic parsing for aggregation questions

   Kun Xu, Sheng Zhang, Yansong Feng, and 2 more authors

   In Proceedings of the AAAI Conference on Artificial Intelligence, Jul 2015

2014

1. Encoding Relation Requirements for Relation Extraction via Joint Inference

   Liwei Chen, Yansong Feng, Songfang Huang, and 2 more authors

   In Proceedings of the 52nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), Jun 2014
2. Joint Inference for Knowledge Base Population

   Liwei Chen, Yansong Feng, Jinghui Mo, and 2 more authors

   In Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing (EMNLP), Oct 2014
3. Answering natural language questions via phrasal semantic parsing

   Kun Xu, Sheng Zhang, Yansong Feng, and 1 more author

   In CCF International Conference on Natural Language Processing and Chinese Computing, Oct 2014
4. Community-based matrix factorization for scalable music recommendation on smartphones

   Jinghui Mo, Yansong Feng, Aixia Jia, and 3 more authors

   In 2014 IEEE International Conference on Multimedia and Expo (ICME), Oct 2014
5. A Robust Audio Similarity Estimation Method for Audio Alignment in Mobile Karaoke Apps

   Jinghui Mo, Yansong Feng, and Dongyan Zhao

   In Proceedings of International Conference on Multimedia Retrieval, Oct 2014

2013

1. Extracting relations from the web via weakly supervised learning

   L Chen, Y Feng, and D Zhao

   Journal of Computer Research and Development, Oct 2013
2. Taxonomy based personalized news recommendation: Novelty and diversity

   Junyang Rao, Aixia Jia, Yansong Feng, and 1 more author

   In International Conference on Web Information Systems Engineering, Oct 2013
3. Efficient simrank-based similarity join over large graphs

   Weiguo Zheng, Lei Zou, Yansong Feng, and 2 more authors

   Proceedings of the VLDB Endowment, Oct 2013
4. S-store: An engine for large rdf graph integrating spatial information

   Dong Wang, Lei Zou, Yansong Feng, and 3 more authors

   In Database Systems for Advanced Applications: 18th International Conference, DASFAA 2013, Wuhan, China, April 22-25, 2013. Proceedings, Part II 18, Oct 2013
5. Personalized news recommendation using ontologies harvested from the web

   Junyang Rao, Aixia Jia, Yansong Feng, and 1 more author

   In International conference on web-age information management, Oct 2013
6. Two-stage multiple kernel learning with multiclass kernel polarization

   Tinghua Wang, Dongyan Zhao, and Yansong Feng

   Knowledge-Based Systems, Oct 2013

2012

1. Explore Person Specific Evidence in Web Person Name Disambiguation

   Liwei Chen, Yansong Feng, Lei Zou, and 1 more author

   In Proceedings of the 2012 Joint Conference on Empirical Methods in Natural Language Processing and Computational Natural Language Learning, Jul 2012
2. Automatic caption generation for news images

   Yansong Feng, and Mirella Lapata

   IEEE transactions on pattern analysis and machine intelligence, Jul 2012

2010

1. Title Generation with Quasi-Synchronous Grammar

   Kristian Woodsend, Yansong Feng, and Mirella Lapata

   In Proceedings of the 2010 Conference on Empirical Methods in Natural Language Processing, Oct 2010
2. Visual Information in Semantic Representation

   Yansong Feng, and Mirella Lapata

   In Human Language Technologies: The 2010 Annual Conference of the North American Chapter of the Association for Computational Linguistics, Jun 2010
3. Topic Models for Image Annotation and Text Illustration

   Yansong Feng, and Mirella Lapata

   In Human Language Technologies: The 2010 Annual Conference of the North American Chapter of the Association for Computational Linguistics, Jun 2010
4. How Many Words Is a Picture Worth? Automatic Caption Generation for News Images

   Yansong Feng, and Mirella Lapata

   In Proceedings of the 48th Annual Meeting of the Association for Computational Linguistics, Jul 2010

2008

1. Automatic Image Annotation Using Auxiliary Text Information

   Yansong Feng, and Mirella Lapata

   In Proceedings of ACL-08: HLT, Jun 2008

2006

1. A novel fingerprint matching scheme based on local structure compatibility

   Yansong Feng, Jufu Feng, Xiaoguang Chen, and 1 more author

   In 18th International Conference on Pattern Recognition (ICPR’06), Jun 2006