查看原文
其他

EMNLP 2022 | 主会长文论文分类整理

杨晨 RUC AI Box 2022-12-14
  © 作者|杨晨
  机构|中国人民大学高瓴人工智能学院 
  研究方向 | 推荐系统 

本文选取了EMNLP 2022中714篇主会长文,参照不同的研究主题进行分类整理(分类标准参考EMNLP官方投稿主题),以供参考。文章也同步发布在AI Box知乎专栏(知乎搜索 AI Box专栏),欢迎大家在知乎专栏的文章下方评论留言,交流探讨! 


导读

EMNLP 2022(The 2022 Conference on Empirical Methods in Naturel Language Processing)是自然语言处理领域的顶级国际会议,由国际语言学会SIGDA小组在世界范围内每年召开一次,计划于2022年12月7日-12月11日以线上线下混合方式在阿联酋阿布扎比举行。EMNLP是CCF推荐的B类国际学术会议,在自然语言处理领域享有很高的学术声誉。本次会议主会共录取长文714篇(其中162篇oral),短文113篇(其中11篇oral);子刊Findings录取长文453篇,短文94篇;还有8篇需要通过道德审查后接收。由于篇幅原因,我们选择了主会的714篇长文进行了分类,其余可以参见如下链接
https://2022.emnlp.org

从词云图看今年EMNLP的研究热点根据长文的标题绘制如下词云图,可以看到今年研究方向主要集中在 Generation、Knowledge、Reasoning 等方向。

目录
1. Commonsense Reasoning【常识推理】
2. Computational Social Science and Cultural Analytics【计算社会科学和文化分析】
3. Dialogue and Interactive Systems【对话和交互系统】
4. Discourse and Pragmatics【语用学】
5. Efficient Methods for NLP【自然语言处理的高效方法】
6. Ethics【伦理】
7. Information Extraction【信息抽取】
8. Information Retrieval and Text Mining【信息检索与文本挖掘】
9. Interpretability, Interactivity and Analysis of Models for NLP【自然语言处理模型的可解释性、交互与分析】
10. Language Modeling and Analysis of Language Models【语言建模与语言模型分析】
11. Linguistic Theories, Cognitive Modeling and Psycholinguistics【语言学理论、认知模型与心理语言学】
12. Machine Learning for NLP【机器学习与自然语言处理】
13. Machine Translation【机器翻译】
14. Multilinguality【多语言】
15. Natural Language Generation【自然语言生成】
16. NLP Applications【自然语言处理应用】
17. Phonology, Morphology and Word Segmentation【音韵、词法和分词】
18. Question Answering【问答系统】
19. Resources and Evaluation【资源和评测】
20. Semantics: Lexical, Sentence level, Textual Inference and Other areas【语义学:单词、句子层次、文本推理和其他领域】
21. Sentiment Analysis, Stylistic Analysis, and Argument Mining【情感分析、文体分析和论点挖掘】
22. Speech, Vision, Robotics, Multimodal Grounding【语音、视觉、机器人、多模态】
23. Summarization【文本摘要】
24. Syntax, Parsing and their Applications【语法、句法及其应用】
25. Theme Track【主题追踪】
26. Unsupervised and Weakly-Supervised Methods in NLP【自然语言处理中的无监督和弱监督方法】


1. Commonsense Reasoning【常识推理】 

  • Identifying Physical Object Use in Sentences
  • A Systematic Investigation of Commonsense Knowledge in Large Language Models
  • Metric-guided Distillation: Distilling Knowledge from the Metric to Ranker
  • and Retriever for Generative Commonsense Reasoning
  • Using Commonsense Knowledge to Answer Why-Questions
  • Maieutic Prompting: Logically Consistent Reasoning with Recursive
  • Explanations
  • Language Models of Code are Few-Shot Commonsense Learners
  • Enhancing Self-Consistency and Performance of Pre-Trained Language
  • Models through Natural Language Inference
  • EvEntS ReaLM: Event Reasoning of Entity States via Language Models
  • GeoMLAMA: Geo-Diverse Commonsense Probing on Multilingual Pre
  • Trained Language Models
  • Memory-assisted prompt editing to improve GPT-3 after deployment
  • Retrieval Augmentation for Commonsense Reasoning: A Unified Approach
  • ReCo: Reliable Causal Chain Reasoning via Structural Causal Recurrent
  • Neural Networks
  • Graph Hawkes Transformer for Extrapolated Reasoning on Temporal Know
  • edge Graphs
  • A Sequential Flow Control Framework for Multi-hop Knowledge Base Que
  • tion Answering
  • ACENet: Attention Guided Commonsense Reasoning on Hybrid Knowledge
  • Graph
  • TranSHER: Translating Knowledge Graph Embedding with Hyper-Ellipsoidal
  • Restriction
  • Rainier: Reinforced Knowledge Introspector for Commonsense Question
  • Answering
  • Understanding ME? Multimodal Evaluation for Fine-grained Visual Commonsense

2. Computational Social Science and Cultural Analytics【计算社会科学和文化分析】 

  • Offer a Different Perspective: Modeling the Belief Alignment of Arguments in Multi-party Debates
  • Prompting for Multimodal Hateful Meme Classification
  • Affective Idiosyncratic Responses to Music
  • Modeling Information Change in Science Communication with Semantically Matched Paraphrases
  • Discovering Differences in the Representation of People using Contextualized Semantic Axes
  • How to disagree well: Investigating the dispute tactics used on Wikipedia
  • Borrowing Human Senses: Comment-Aware Self-Training for Social Media Multimodal Classification
  • Unifying Data Perspectivism and Personalization: An Application to Social Norms
  • Empowering the Fact-checkers! Automatic Identification of Claim Spans on Twitter
  • Sequence Models for Document Structure Identification in an Undeciphered Script
  • Distilling Context for Toxicity Detection via Information Bottleneck
  • Sentence-level Media Bias Analysis Informed by Discourse Structures


3. Dialogue and Interactive Systems【对话和交互系统】 

  • Navigating Connected Memories with a Task-oriented Dialog System
  • FETA: A Benchmark for Few-Sample Task Transfer in Open-Domain Dialogue
  • IM^2: an Interpretable and Multi-category Integrated Metric Framework for Automatic Dialogue Evaluation
  • Prompt Conditioned VAE: Enhancing Generative Replay for Lifelong Learning in Task-Oriented Dialogue
  • End-to-End Neural Discourse Deixis Resolution in Dialogue
  • CDialog: A Multi-turn Covid-19 Conversation Dataset for Entity-Aware Dialog Generation
  • When More Data Hurts: A Troubling Quirk in Developing Broad-Coverage Natural Language Understanding Systems
  • Injecting Domain Knowledge in Language Models for Task-oriented Dialogue Systems
  • DialogConv: A Lightweight Fully Convolutional Network for Multi-view Response Selection
  • CDConv: A Benchmark for Contradiction Detection in Chinese Conversations
  • Co-guiding Net: Achieving Mutual Guidances between Multiple Intent Detection and Slot Filling via Heterogeneous Semantics-Label Graphs
  • Estimating Soft Labels for Out-of-Domain Intent Detection
  • InstructDial: Improving Zero and Few-shot Generalization in Dialogue through Instruction Tuning
  • Aligning Recommendation and Conversation via Dual Imitation
  • Correctable-DST: Mitigating Historical Context Mismatch between Training and Inference for Improved Dialogue State Tracking
  • MetaASSIST: Robust Dialogue State Tracking with Meta Learning
  • Watch the Neighbors: A Unified K-Nearest Neighbor Contrastive Learning Framework for OOD Intent Discovery
  • Counterfactual Data Augmentation via Perspective Transition for Open-Domain Dialogues
  • There Is No Standard Answer: Knowledge-Grounded Dialogue Generation with Adversarial Activated Multi-Reference Learning
  • Back to the Future: Bidirectional Information Decoupling Network for Multi-turn Dialogue Modeling
  • Improving Multi-turn Emotional Support Dialogue Generation with Lookahead Strategy Planning
  • Robots-Dont-Cry: Understanding Falsely Anthropomorphic Utterances in Dialog Systems
  • FineD-Eval: Fine-grained Automatic Dialogue-Level Evaluation
  • Group is better than individual: Exploiting Label Topologies and Label Relations for Joint Multiple Intent Detection and Slot Filling
  • ProsocialDialog: A Prosocial Backbone for Conversational Agents
  • CGoDial: A Large-Scale Benchmark for Chinese Goal-oriented Dialog Evaluation
  • Information-Theoretic Text Hallucination Reduction for Video-grounded Dialogue
  • STRUDEL: Structured Dialogue Summarization for Dialogue Comprehension
  • BotsTalk: Machine-sourced Framework for Automatic Curation of Large-scale Multi-skill Dialogue Datasets
  • Supervised Prototypical Contrastive Learning for Emotion Recognition in Conversation
  • META-GUI: Towards Multi-modal Conversational Agents on Mobile GUI
  • Q-TOD: A Query-driven Task-oriented Dialogue System
  • dial2vec: Self-Guided Contrastive Learning of Unsupervised Dialogue Embeddings
  • Multi-Label Intent Detection via Contrastive Task Specialization of Sentence Encoders
  • Enhancing Joint Multiple Intent Detection and Slot Filling with Global Intent-Slot Co-occurrence
  • IRRGN: An Implicit Relational Reasoning Graph Network for Multi-turn Response Selection
  • Dungeons and Dragons as a Dialog Challenge for Artificial Intelligence
  • CONQRR: Conversational Query Rewriting for Retrieval with Reinforcement Learning
  • Towards Efficient Dialogue Pre-training with Transferable and Interpretable Latent Structure
  • "Reflect, Not Reflex: Inference-Based Common Ground Improves Dialogue Response Quality"
  • FlowEval: A Consensus-Based Dialogue Evaluation Framework Using Segment Act Flows
  • Eliciting Knowledge from Large Pre-Trained Models for Unsupervised Knowledge-Grounded Conversation
  • Structural Constraints and Natural Language Inference for End-to-End Flowchart Grounded Dialog Response Generation

4. Discourse and Pragmatics【语用学】 

  • Discourse Comprehension: A Question Answering Framework to Represent Sentence Connections
  • Sentence-Incremental Neural Coreference Resolution
  • Concadia: Towards Image-Based Text Generation with a Purpose
  • "An Unsupervised, Geometric and Syntax-aware Quantification of Polysemy"
  • Facilitating Contrastive Learning of Discourse Relational Senses by Exploiting the Hierarchy of Sense Relations

5. Efficient Methods for NLP【自然语言处理的高效方法】 

  • Training Dynamics for Curriculum Learning: A Study on Monolingual and Cross-lingual NLU
  • Model Cascading: Towards Jointly Improving Efficiency and Accuracy of NLP Systems
  • "LightEA: A Scalable, Robust, and Interpretable Entity Alignment Framework via Three-view Label Propagation"
  • VIRT: Improving Representation-based Text Matching via Virtual Interaction
  • Learning Label Modular Prompts for Text Classification in the Wild
  • COST-EFF: Collaborative Optimization of Spatial and Temporal Efficiency with Slenderized Multi-exit Language Models
  • Neural-based Mixture Probabilistic Query Embedding for Answering FOL queries on Knowledge Graphs
  • Sparse Teachers Can Be Dense with Knowledge
  • The Optimal BERT Surgeon: Scalable and Accurate Second-Order Pruning for Large Language Models
  • An Efficient Memory-Augmented Transformer for Knowledge-Intensive NLP Tasks
  • LittleBird: Efficient Faster & Longer Transformer for Question Answering
  • Understanding and Improving Knowledge Distillation for Quantization Aware Training of Large Transformer Encoders
  • Vector-Quantized Input-Contextualized Soft Prompts for Natural Language Understanding
  • Tutoring Helps Students Learn Better: Improving Knowledge Distillation for BERT with Tutor Network
  • Efficient Pre-training of Masked Language Model via Concept-based Curriculum Masking
  • HashFormers: Towards Vocabulary-independent Pre-trained Transformers
  • Calibrating Student Models for Emotion-related Tasks
  • Overcoming Catastrophic Forgetting in Zero-Shot Cross-Lingual Generation
  • Leveraging QA Datasets to Improve Generative Data Augmentation
  • EdgeFormer: A Parameter-Efficient Transformer for On-Device Seq2seq Generation

6. Ethics【伦理】 

  • ""I'm sorry to hear that"": Finding New Biases in Language Models with a Holistic Descriptor Dataset"
  • Exploration of the Usage of Color Terms by Color-blind Participants in Online Discussion Platforms
  • ArtELingo: A Million Emotion Annotations of WikiArt with Emphasis on Diversity over Language and Culture
  • PAIR: Prompt-Aware margIn Ranking for Counselor Reflection Scoring in Motivational Interviewing

  • NewsClaims: A New Benchmark for Claim Detection from News with Attribute Knowledge

  • POQue: Asking Participant-specific Outcome Questions for a Deeper Understanding of Complex Events

  • MEE: A Novel Multilingual Event Extraction Dataset

  • Late Fusion with Triplet Margin Objective for Multimodal Ideology Prediction and Analysis

  • Do Vision-and-Language Transformers Learn Grounded Predicate-Noun Dependencies?

  • MuRAG: Multimodal Retrieval-Augmented Generator for Open Question Answering over Images and Text

  • Differentiable Data Augmentation for Contrastive Sentence Representation Learning

  • Balancing out Bias: Achieving Fairness Through Balanced Training

  • COLD: A Benchmark for Chinese Offensive Language Detection

  • Gendered Mental Health Stigma in Masked Language Models

  • SafeText: A Benchmark for Exploring Physical Safety in Language Models

  • BERTScore is Unfair: On Social Bias in Language Model-Based Metrics for Text Generation

  • Just Fine-tune Twice: Selective Differential Privacy for Large Language Models

  • TextFusion: Privacy-Preserving Pre-trained Model Inference via Token Fusion

  • Perturbation Augmentation for Fairer NLP

  • Debiasing Pretrained Text Encoders by Paying Attention to Paying Attention

  • MABEL: Attenuating Gender Bias using Textual Entailment Data


7. Information Extraction【信息抽取】 

  • Transfer Learning from Semantic Role Labeling to Event Argument Extraction with Template-based Slot Querying
  • Generative Knowledge Graph Construction: A Review
  • Graph-based Model Generation for Few-Shot Relation Extraction
  • "A Good Neighbor, A Found Treasure: Mining Treasured Neighbors for Knowledge Graph Entity Typing"
  • ReSel: N-ary Relation Extraction from Scientific Text and Tables by Learning to Retrieve and Select
  • "MAVEN-ERE: A Unified Large-scale Dataset for Event Coreference, Temporal, Causal, and Subevent Relation Extraction"
  • Entity Extraction in Low Resource Domains with Selective Pre-training of Large Language Models
  • Multilingual Relation Classification via Efficient and Effective Prompting
  • Fine-grained Contrastive Learning for Relation Extraction
  • SQUIRE: A Sequence-to-sequence Framework for Multi-hop Knowledge Graph Reasoning
  • Style Transfer as Data Augmentation: A Case Study on Named Entity Recognition
  • SetGNER: General Named Entity Recognition as Entity Set Generation
  • SpanProto: A Two-stage Span-based Prototypical Network for Few-shot Named Entity Recognition
  • Logical Neural Networks for Knowledge Base Completion with Embeddings & Rules
  • Syntactic Multi-view Learning for Open Information Extraction
  • A Unified Positive-Unlabeled Learning Framework for Document-Level Relation Extraction with Different Levels of Labeling
  • RelU-Net: Syntax-aware Graph U-Net for Relational Triple Extraction
  • Retrieval-Augmented Generative Question Answering for Event Argument Extraction
  • Wider & Closer: Mixture of Short-channel Distillers for Zero-shot Cross-lingual Named Entity Recognition
  • Syntactically Rich Discriminative Training: An Effective Method for Open Information Extraction
  • Simple Questions Generate Named Entity Recognition Datasets
  • Bi-Directional Iterative Prompt-Tuning for Event Argument Extraction
  • Learning Robust Representations for Continual Relation Extraction via Adversarial Class Augmentation
  • Attention and Edge-Label Guided Graph Convolutional Networks for Named Entity Recognition
  • Improving Event Coreference Resolution Using Document-level and Topic-level Information
  • Modeling Label Correlations for Ultra-Fine Entity Typing with Neural Pairwise Conditional Random Field
  • Open Relation and Event Type Discovery with Type Abstraction
  • Cross-stitching Text and Knowledge Graph Encoders for Distantly Supervised Relation Extraction
  • Better Few-Shot Relation Extraction with Label Prompt Dropout
  • UniRel: Unified Representation and Interaction for Joint Relational Triple Extraction
  • MetaTKG: Learning Evolutionary Meta-Knowledge for Temporal Knowledge Graph Reasoning
  • WR-One2Set: Towards Well-Calibrated Keyphrase Generation
  • Query-based Instance Discrimination Network for Relational Triple Extraction
  • MatchPrompt: Prompt-based Open Relation Extraction with Semantic Consistency Guided Clustering
  • Towards Better Document-level Relation Extraction via Iterative Inference
  • IELM: An Open Information Extraction Benchmark for Pre-Trained Language Models
  • EDIN: An End-to-end Benchmark and Pipeline for Unknown Entity Discovery and Indexing
  • "Learning Cross-Task Dependencies for Joint Extraction of Entities, Events, Event Arguments, and Relations"
  • Entity-centered Cross-document Relation Extraction
  • Boosting Document-Level Relation Extraction by Mining and Injecting Logical Rules
  • """Covid vaccine is against Covid but Oxford vaccine is made at Oxford!"" Semantic Interpretation of Proper Noun Compounds"
  • Towards relation extraction from speech

8. Information Retrieval and Text Mining【信息检索与文本挖掘】 

  • Topic Modeling With Topological Data Analysis
  • WeDef: Weakly Supervised Backdoor Defense for Text Classification
  • Pseudo-Relevance for Enhancing Document Representation
  • Certified Error Control of Candidate Set Pruning for Two-Stage Relevance Ranking
  • RetroMAE: Pre-Training Retrieval-oriented Language Models Via Masked Auto-Encoder
  • Efficient Document Retrieval by End-to-End Refining and Quantizing BERT Embedding with Contrastive Product Quantization
  • Prompt-Based Meta-Learning For Few-shot Text Classification
  • Generative Multi-hop Retrieval
  • COCO-DR: Combating the Distribution Shift in Zero-Shot Dense Retrieval with Contrastive and Distributionally Robust Learning
  • Mitigating Data Sparsity for Short Text Topic Modeling by Topic-Semantic Contrastive Learning
  • CodeRetriever: A Large Scale Contrastive Pre-Training Method for Code Search
  • ConvTrans: Transforming Web Search Sessions for Conversational Dense Retrieval
  • Improving Multi-task Stance Detection with Multi-task Interaction Network
  • Generative Entity Typing with Curriculum Learning
  • Towards Reinterpreting Neural Topic Models via Composite Activations
  • Explicit Query Rewriting for Conversational Dense Retrieval
  • Exploring Representation-level Augmentation for Code Search
  • DuReader-Retrieval: A Large-scale Chinese Benchmark for Passage Retrieval from Web Search Engine
  • OTSeq2Set: An Optimal Transport Enhanced Sequence-to-Set Model for Extreme Multi-label Text Classification
  • A Framework for Adapting Pre-Trained Language Models to Knowledge Graph Completion
  • Reduce Catastrophic Forgetting of Dense Retrieval Training with Teleportation Negatives
  • An Adaptive Logical Rule Embedding Model for Inductive Reasoning over Temporal Knowledge Graphs
  • A Unified Neural Network Model for Readability Assessment with Feature Projection and Length-Balanced Loss
  • Incorporating Relevance Feedback for Information-Seeking Retrieval using Few-Shot Document Re-Ranking
  • Coordinated Topic Modeling
  • Large Dual Encoders Are Generalizable Retrievers
  • CODER: An efficient framework for improving retrieval through COntextual Document Embedding Reranking
  • Recovering Gold from Black Sand: Multilingual Dense Passage Retrieval with Hard and False Negative Samples

9. Interpretability, Interactivity and Analysis of Models for NLP【自然语言处理模型的可解释性、交互与分析】 

  • Revisiting Parameter-Efficient Tuning: Are We Really There Yet?
  • Can Transformers Reason in Fragments of Natural Language?
  • Why Should Adversarial Perturbations be Imperceptible? Rethink the Research Paradigm in Adversarial NLP
  • Predicting Fine-Tuning Performance with Probing
  • TASA: Deceiving Question Answering Models by Twin Answer Sentences Attack
  • Transformer Feed-Forward Layers Build Predictions by Promoting Concepts in the Vocabulary Space
  • Interpreting Language Models with Contrastive Explanations
  • Balanced Adversarial Training: Balancing Tradeoffs between Fickleness and Obstinacy in NLP Models
  • DropMix: A Textual Data Augmentation Combining Dropout with Mixup
  • """Will You Find These Shortcuts?"" A Protocol for Evaluating the Faithfulness of Input Salience Methods for Text Classification"
  • On the Transformation of Latent Space in Fine-Tuned NLP Models
  • A Multilingual Perspective Towards the Evaluation of Attribution Methods in Natural Language Inference
  • Robustness of Demonstration-based Learning Under Limited Data Scenario
  • Entailer: Answering Questions with Faithful and Truthful Chains of Reasoning
  • That's the Wrong Lung! Evaluating and Improving the Interpretability of Unsupervised Multimodal Encoders for Medical Data
  • Logical Reasoning with Span-Level Predictions for Interpretable and Robust NLI Models
  • Finding Dataset Shortcuts with Grammar Induction
  • SLING: Sino Linguistic Evaluation of Large Language Models
  • Towards Interactivity and Interpretability: A Rationale-based Legal Judgment Prediction Framework
  • Adversarial Concept Erasure in Kernel Space
  • ADDMU: Detection of Far-Boundary Adversarial Examples with Data and Model Uncertainty Estimation
  • Let the CAT out of the bag: Contrastive Attributed explanations for Text
  • Does Self-Rationalization Improve Robustness to Spurious Correlations?
  • Cross-Linguistic Syntactic Difference in Multilingual BERT: How Good is It and How Does It Affect Transfer?
  • Efficient Adversarial Training with Robust Early-Bird Tickets
  • Learning to Explain Selectively: A Case Study on Question Answering
  • Better Hit the Nail on the Head than Beat around the Bush: Removing Protected Attributes with a Single Projection
  • Does Your Model Classify Entities Reasonably? Diagnosing and Mitigating Spurious Correlations in Entity Typing
  • Measuring the Mixing of Contextual Information in the Transformer
  • Decoding a Neural Retriever's Latent Space for Query Suggestion
  • Towards Teachable Reasoning Systems: Using a Dynamic Memory of User Feedback for Continual System Improvement
  • Are All Spurious Features in Natural Language Alike? An Analysis through a Causal Lens
  • Human Guided Exploitation of Interpretable Attention Patterns in Summarization and Topic Segmentation

10. Language Modeling and Analysis of Language Models【语言建模与语言模型分析】 

  • Iteratively Prompt Pre-trained Language Models for Chain of Thought
  • XPrompt: Exploring the Extreme of Prompt Tuning
  • Rethinking the Role of Demonstrations: What Makes In-Context Learning Work?
  • Finding Skill Neurons in Pre-trained Transformer-based Language Models
  • Instance Regularization for Discriminative Language Model Pre-training
  • ZeroGen: Efficient Zero-shot Learning via Dataset Generation
  • Efficient Large Scale Language Modeling with Mixtures of Experts
  • Model Criticism for Long-Form Text Generation
  • The Geometry of Multilingual Language Model Representations
  • What Makes Instruction Learning Hard? An Investigation and a New Challenge in a Synthetic Environment
  • Language Model Pre-Training with Sparse Latent Typing
  • Ground-Truth Labels Matter: A Deeper Look into Input-Label Demonstrations
  • ROSE: Robust Selective Fine-tuning for Pre-trained Language Models
  • Revisiting Pre-trained Language Models and their Evaluation for Arabic Natural Language Processing
  • Knowledge Prompting in Pre-trained Language Model for Natural Language Understanding
  • Nearest Neighbor Zero-Shot Inference
  • Red Teaming Language Models with Language Models
  • COPEN: Probing Conceptual Knowledge in Pre-trained Language Models
  • Training Language Models with Memory Augmentation
  • Invariant Language Modeling
  • AdaMix: Mixture-of-Adaptations for Parameter-efficient Model Tuning
  • BioReader: a Retrieval-Enhanced Text-to-Text Transformer for Biomedical Literature
  • InforMask: Unsupervised Informative Masking for Language Model Pretraining
  • Fine-tuned Language Models are Continual Learners
  • TemporalWiki: A Lifelong Benchmark for Training and Evaluating Ever-Evolving Language Models
  • Improving Temporal Generalization of Pre-trained Language Models with Lexical Semantic Change
  • ATTEMPT: Parameter-Efficient Multi-task Tuning via Attentional Mixtures of Soft Prompts
  • Exploring Mode Connectivity for Pre-trained Language Models
  • Boosting Natural Language Generation from Instructions with Meta-Learning
  • Zero-Shot Learners for Natural Language Understanding via a Unified Multiple Choice Perspective
  • Subword Evenness (SuE) as a Predictor of Cross-lingual Transfer to Low-resource Languages
  • Parameter-Efficient Tuning Makes a Good Classification Head
  • Character-level White-Box Adversarial Attacks against Transformers via Attachable Subwords Substitution
  • "SocioProbe: What, When, and Where Language Models Learn about Sociodemographics"
  • GPS: Genetic Prompt Search for Efficient Few-Shot Learning
  • Active Example Selection for In-Context Learning
  • Adapting a Language Model While Preserving its General Knowledge

11. Linguistic Theories, Cognitive Modeling and Psycholinguistics【语言学理论、认知模型与心理语言学】 

  • The Curious Case of Control
  • Tracing Semantic Variation in Slang
  • A Comprehensive Comparison of Neural Networks as Cognitive Models of Inflection
  • Cascading Biases: Investigating the Effect of Heuristic Annotation Strategies on Data and Models
  • Is the Brain Mechanism for Hierarchical Structure Building Universal Across Languages? An fMRI Study of Chinese and English
  • Entropy- and Distance-Based Predictors From GPT-2 Attention Patterns Predict Reading Times Over and Above GPT-2 Surprisal
  • Discourse Context Predictability Effects in Hindi Word Order
  • Context Limitations Make Neural Language Models More Human-Like
  • "The better your Syntax, the better your Semantics? Probing Pretrained Language Models for the English Comparative Correlative"


12. Machine Learning for NLP【机器学习与自然语言处理】 

  • Diverse Parallel Data Synthesis for Cross-Database Adaptation of Text-to-SQL Parsers
  • Fixing Model Bugs with Natural Language Patches
  • Interventional Training for Out-Of-Distribution Natural Language Understanding
  • Backdoor Attacks in Federated Learning by Rare Embeddings and Gradient Ensembling
  • When Can Transformers Ground and Compose: Insights from Compositional Generalization Benchmarks
  • GammaE: Gamma Embeddings for Logical Queries on Knowledge Graphs
  • Inducer-tuning: Connecting Prefix-tuning and Adapter-tuning
  • Numerical Optimizations for Weighted Low-rank Estimation on Language Models
  • Efficient Nearest Neighbor Search for Cross-Encoder Models using Matrix Factorization
  • A Localized Geometric Method to Match Knowledge in Low-dimensional Hyperbolic Space
  • Making Pretrained Language Models Good Long-tailed Learners
  • RLPrompt: Optimizing Discrete Text Prompts with Reinforcement Learning
  • Natural Language to Code Translation with Execution
  • HPT: Hierarchy-aware Prompt Tuning for Hierarchical Text Classification
  • GA-SAM: Gradient-Strength based Adaptive Sharpness-Aware Minimization for Improved Generalization
  • BBTv2: Towards a Gradient-Free Future with Large Language Models
  • Passage-Mask: A Learnable Regularization Strategy for Retriever-Reader Models
  • Mixture of Attention Heads: Selecting Attention Heads Per Token
  • Complex Hyperbolic Knowledge Graph Embeddings with Fast Fourier Transform
  • Transformer-based Entity Typing in Knowledge Graphs
  • A Survey of Active Learning for Natural Language Processing
  • G-MAP: General Memory-Augmented Pre-trained Language Model for Domain Tasks
  • Textual Manifold-based Defense Against Natural Language Adversarial Examples
  • The Devil in Linear Transformer
  • STGN: an Implicit Regularization Method for Learning with Noisy Labels in Natural Language Processing
  • Learning Inter-Entity-Interaction for Few-Shot Knowledge Graph Completion
  • Hierarchical Phrase-Based Sequence-to-Sequence Learning
  • Adaptive Label Smoothing with Self-Knowledge in Natural Language Generation
  • Variational Autoencoder with Disentanglement Priors for Low-Resource Task-Specific Natural Language Generation
  • MM-Align: Learning Optimal Transport-based Alignment Dynamics for Fast and Accurate Inference on Missing Modality Sequences

13. Machine Translation【机器翻译】 

  • Sampling-Based Approximations to Minimum Bayes Risk Decoding for Neural Machine Translation
  • PreQuEL: Quality Estimation of Machine Translation Outputs in Advance
  • Digging Errors in NMT: Evaluating and Understanding Model Errors from Partial Hypothesis Space
  • The Importance of Being Parameters: An Intra-Distillation Method for Serious Gains
  • Non-Parametric Domain Adaptation for End-to-End Speech Translation
  • Information-Transport-based Policy for Simultaneous Translation
  • Multilingual Machine Translation with Hyper-Adapters
  • Continual Learning of Neural Machine Translation within Low Forgetting Risk Regions
  • Distill The Image to Nowhere: Inversion Knowledge Distillation for Multimodal Machine Translation
  • Machine Translation Robustness to Natural Asemantic Variation
  • Neural Machine Translation with Contrastive Translation Memories
  • A Template-based Method for Constrained Neural Machine Translation
  • Chunk-based Nearest Neighbor Machine Translation
  • MT-GenEval: A Counterfactual and Contextual Dataset for Evaluating Gender Accuracy in Machine Translation
  • Candidate Soups: Fusing Candidate Results Improves Translation Quality for Non-Autoregressive Translation
  • Competency-Aware Neural Machine Translation: Can Machine Translation Know its Own Translation Quality?
  • Multi-Granularity Optimization for Non-Autoregressive Translation
  • WeTS: A Benchmark for Translation Suggestion
  • Norm-based Noisy Corpora Filtering and Refurbishing in Neural Machine Translation
  • Towards Robust k-Nearest-Neighbor Machine Translation
  • SimQA: Detecting Simultaneous MT Errors through Word-by-Word Question Answering
  • Modeling Consistency Preference via Lexical Chains for Document-level Neural Machine Translation
  • Breaking the Representation Bottleneck of Chinese Characters: Neural Machine Translation with Stroke Sequence Modeling
  • Increasing Visual Awareness in Multimodal Neural Machine Translation from an Information Theoretic Perspective
  • Unifying the Convergences in Multilingual Neural Machine Translation
  • XLM-D: Decorate Cross-lingual Pre-training Model as Non-Autoregressive Neural Machine Translation
  • Hypoformer: Hybrid Decomposition Transformer for Edge-friendly Neural Machine Translation
  • When does Parameter-Efficient Transfer Learning Work for Machine Translation?
  • Bilingual Synchronization: Restoring Translational Relationships with Editing Operations
  • ConsistTL: Modeling Consistency in Transfer Learning for Low-Resource Neural Machine Translation
  • Multimodal Robustness for Neural Machine Translation
  • Disentangling Uncertainty in Machine Translation Evaluation
  • Towards Opening the Black Box of Neural Machine Translation: Source and Target Interpretations of the Transformer
  • RAPO: An Adaptive Ranking Paradigm for Bilingual Lexicon Induction
  • Adaptive Token-level Cross-lingual Feature Mixing for Multilingual Neural Machine Translation
  • Low-resource Neural Machine Translation with Cross-modal Alignment
  • Entropy-Based Vocabulary Substitution for Incremental Learning in Multilingual Neural Machine Translation

14. Multilinguality【多语言】 

  • Calibrating Zero-shot Cross-lingual (Un-)structured Predictions
  • Zero-shot Cross-lingual Transfer of Prompt-based Tuning with a Unified Multilingual Prompt
  • Transforming Sequence Tagging Into A Seq2Seq Task
  • Joint Completion and Alignment of Multilingual Knowledge Graphs
  • Improving Low-Resource Languages in Pre-Trained Multilingual Language Models
  • Crossmodal-3600: A Massively Multilingual Multimodal Evaluation Dataset
  • Graph-Based Multilingual Label Propagation for Low-Resource Part-of-Speech Tagging
  • AfroLID: A Neural Language Identification Tool for African Languages
  • The (Undesired) Attenuation of Human Biases by Multilinguality
  • CoCoA: An Encoder-Decoder Model for Controllable Code-switched Generation
  • Multi-level Distillation of Semantic Knowledge for Pre-training Multilingual Language Model
  • Empowering Dual-Encoder with Query Generator for Cross-Lingual Dense Retrieval
  • Analyzing the Mono- and Cross-Lingual Pretraining Dynamics of Multilingual Language Models
  • Cross-Align: Modeling Deep Cross-lingual Interactions for Word Alignment
  • Discovering Low-rank Subspaces for Language-agnostic Multilingual Representations
  • Data-Efficient Strategies for Expanding Hate Speech Detection into Under-Resourced Languages
  • T-Modules: Translation Modules for Zero-Shot Cross-Modal Machine Translation
  • Enhancing Multilingual Language Model with Massive Multilingual Knowledge Triples
  • Discovering Language-neutral Sub-networks in Multilingual Language Models
  • Hyper-X: A Unified Hypernetwork for Multi-Task Multilingual Transfer
  • ConNER: Consistency Training for Cross-lingual Named Entity Recognition
  • Few-shot Learning with Multilingual Generative Language Models
  • English Contrastive Learning Can Learn Universal Cross-lingual Sentence Embeddings
  • Label-aware Multi-level Contrastive Learning for Cross-lingual Spoken Language Understanding
  • Polyglot Prompt: Multilingual Multitask Prompt Training
  • PRO-CS : An Instance-Based Prompt Composition Technique for Code-Switched Tasks
  • Don't Stop Fine-Tuning: On Training Regimes for Few-Shot Cross-Lingual Transfer with Multilingual Language Models

15. Natural Language Generation【自然语言生成】 

  • RankGen: Improving Text Generation with Large Ranking Models
  • Linearizing Transformer with Key-Value Memory
  • A Unified Encoder-Decoder Framework with Entity Memory
  • A Distributional Lens for Multi-Aspect Controllable Text Generation
  • ELMER: A Non-Autoregressive Pre-trained Language Model for Efficient and Effective Text Generation
  • Curriculum Prompt Learning with Self-Training for Abstractive Dialogue Summarization
  • Visual Spatial Description: Controlled Spatial-Oriented Image-to-Text Generation
  • SubeventWriter: Iterative Sub-event Sequence Generation with Coherence Controller
  • Towards a Unified Multi-Dimensional Evaluator for Text Generation
  • Prompt-and-Rerank: A Method for Zero-Shot and Few-Shot Arbitrary Textual Style Transfer with Small Language Models
  • Gradient-based Constrained Sampling from Language Models
  • DisCup: Discriminator Cooperative Unlikelihood Prompt-tuning for Controllable Text Generation
  • CapOnImage: Context-driven Dense-Captioning on Image
  • DSM: Question Generation over Knowledge Base via Modeling Diverse Subgraphs with Meta-learner
  • Re3: Generating Longer Stories With Recursive Reprompting and Revision
  • Discourse-Aware Soft Prompting for Text Generation
  • Context-Situated Pun Generation
  • Differentially Private Language Models for Secure Data Sharing
  • Conditional set generation using Seq2seq models
  • Twist Decoding: Diverse Generators Guide Each Other
  • Contrastive Learning enhanced Author-Style Headline Generation
  • IndicNLG Benchmark: Multilingual Datasets for Diverse NLG Tasks in Indic Languages
  • Investigating the Robustness of Natural Language Generation from Logical Forms via Counterfactual Samples
  • PLOG: Table-to-Logic Pretraining for Logical Table-to-Text Generation
  • Controlled Text Reduction
  • Help me write a Poem - Instruction Tuning as a Vehicle for Collaborative Poetry Writing
  • Revisiting Grammatical Error Correction Evaluation and Beyond
  • R2D2: Robust Data-to-Text with Replacement Detection
  • Precisely the Point: Adversarial Augmentations for Faithful and Informative Text Generation
  • Composing Ci with Reinforced Non-autoregressive Text Generation
  • Keyphrase Generation via Soft and Hard Semantic Corrections
  • JANUS: Joint Autoregressive and Non-autoregressive Training with Auxiliary Loss for Sequence Generation
  • Towards Table-to-Text Generation with Pretrained Language Model: A Table Structure Understanding and Text Deliberating Approach
  • T-STAR: Truthful Style Transfer using AMR Graph as Intermediate Representation
  • Towards Inter-character Relationship-driven Story Generation
  • Hard Gate Knowledge Distillation - Leverage Calibration for Robust and Reliable Language Model
  • VisToT: Vision-Augmented Table-to-Text Generation
  • Improving Iterative Text Revision by Learning Where to Edit from Other Revision Tasks
  • Evade the Trap of Mediocrity: Promoting Diversity and Novelty in Text Generation via Concentrating Attention
  • ProofInfer: Generating Proof via Iterative Hierarchical Inference


16. NLP Applications【自然语言处理应用】 

  • Varifocal Question Generation for Fact-checking
  • ConReader: Exploring Implicit Relations in Contracts for Contract Clause Extraction
  • SHARE: a System for Hierarchical Assistive Recipe Editing
  • ScienceWorld: Is your Agent Smarter than a 5th Grader?
  • Federated Meta-Learning for Emotion and Sentiment Aware Multi-modal Complaint Identification
  • Data-Efficient Playlist Captioning With Musical and Linguistic Knowledge
  • Improved grammatical error correction by ranking elementary edits
  • Neighborhood Contrastive Learning for Scientific Document Representations with Citation Embeddings
  • MedJEx: A Medical Jargon Extraction Model with Wiki's Hyperlink Span and Contextualized Masked Language Model Score
  • PAR: Political Actor Representation Learning with Social Context and Expert Knowledge
  • Learning to Generate Question by Asking Question: A Primal-Dual Approach with Uncommon Word Generation
  • Translation between Molecules and Natural Language
  • Guiding Neural Entity Alignment with Compatibility
  • Segmenting Numerical Substitution Ciphers
  • How Large Language Models are Transforming Machine-Paraphrase Plagiarism
  • Deconfounding Legal Judgment Prediction for European Court of Human Rights Cases Towards Better Alignment with Experts
  • PLM-based World Models for Text-based Games
  • Large language models are few-shot clinical information extractors
  • ToKen: Task Decomposition and Knowledge Infusion for Few-Shot Hate Speech Detection
  • PromptEHR: Conditional Electronic Healthcare Records Generation with Prompt Learning
  • Open-Topic False Information Detection on Social Networks with Contrastive Adversarial Learning
  • Rethinking Positional Encoding in Tree Transformer for Code Representation
  • A Joint Learning Framework for Restaurant Survival Prediction and Explanation
  • Life is a Circus and We are the Clowns: Automatically Finding Analogies between Situations and Processes
  • Distilling Causal Effect from Miscellaneous Other-Class for Continual Named Entity Recognition
  • Chapter Ordering in Novels
  • Open-ended Knowledge Tracing for Computer Science Education
  • MedCLIP: Contrastive Learning from Unpaired Medical Images and Text
  • Automatic Generation of Socratic Subquestions for Teaching Math Word Problems
  • Improving Chinese Spelling Check by Character Pronunciation Prediction: The Effects of Adaptivity and Granularity
  • A Speaker-Aware Co-Attention Framework for Medical Dialogue Information Extraction
  • Towards Multi-Modal Sarcasm Detection via Hierarchical Congruity Modeling with Knowledge Enhancement
  • MetaFill: Text Infilling for Meta-Path Generation on Heterogeneous Information Networks
  • Affective Knowledge Enhanced Multiple-Graph Fusion Networks for Aspect-based Sentiment Analysis
  • TeleMelody: Lyric-to-Melody Generation with a Template-Based Two-Stage Method
  • SEEN: Structured Event Enhancement Network for Explainable Need Detection of Information Recall Assistance
  • Tiny-NewsRec: Effective and Efficient PLM-based News Recommendation
  • BERT in Plutarch's Shadows
  • Natural Logic-guided Autoregressive Multi-hop Document Retrieval for Fact Verification
  • Federated Model Decomposition with Private Vocabulary for Text Classification
  • Boundary-Driven Table-Filling for Aspect Sentiment Triplet Extraction
  • Topical Segmentation of Spoken Narratives: A Test Case on Holocaust Survivor Testimonies
  • Factual Accuracy is not Enough: Planning Consistent Description Order for Radiology Report Generation
  • GREENER: Graph Neural Networks for News Media Profiling
  • Cross-lingual neural fuzzy matching for exploiting target-language monolingual corpora in computer-aided translation
  • FormLM: Recommending Creation Ideas for Online Forms by Modelling Semantic and Structural Information
  • Multitask Instruction-based Prompting for Fallacy Recognition
  • Semantic Novelty Detection and Characterization in Factual Text Involving Named Entities
  • Unsupervised Non-transferable Text Classification
  • Mask the Correct Tokens: An Embarrassingly Simple Approach for Error Correction
  • MoSE: Modality Split and Ensemble for Multimodal Knowledge Graph Completion

17. Phonology, Morphology and Word Segmentation【音韵、词法和分词】 

  • Improving Tokenisation by Alternative Treatment of Spaces
  • Unsupervised Boundary-Aware Language Model Pretraining for Chinese Sequence Labeling
  • "Break it Down into BTS: Basic, Tiniest Subword Units for Korean"

18. Question Answering【问答系统】 

  • Pre-training Language Models with Deterministic Factual Knowledge
  • OpenCQA: Open-ended Question Answering with Charts
  • Generating Natural Language Proofs with Verifier-Guided Search
  • Improving Complex Knowledge Base Question Answering via Question-to-Action and Question-to-Question Alignment
  • Successive Prompting for Decomposing Complex Questions
  • M3: A Multi-View Fusion and Multi-Decoding Network for Multi-Document Reading Comprehension
  • Semantic Framework based Query Generation for Temporal Question Answering over Knowledge Graphs
  • Improving compositional generalization for multi-step quantitative reasoning in question answering
  • Learning to Decompose: Hypothetical Question Decomposition Based on Comparable Texts
  • TaCube: Pre-computing Data Cubes for Answering Numerical-Reasoning Questions over Tabular Data
  • Rich Knowledge Sources Bring Complex Knowledge Conflicts: Recalibrating Models to Reflect Conflicting Evidence
  • QA Domain Adaptation using Hidden Space Augmentation and Self-Supervised Contrastive Adaptation
  • Retrieval as Attention: End-to-end Learning of Retrieval and Reading within a Single Transformer
  • Generating Information-Seeking Conversations from Unlabeled Documents
  • You Only Need One Model for Open-domain Question Answering
  • KECP: Knowledge Enhanced Contrastive Prompting for Few-shot Extractive Question Answering
  • Improving Passage Retrieval with Zero-Shot Question Generation
  • FiE: Building a Global Probability Space by Leveraging Early Fusion in Encoder for Open-Domain Question Answering
  • Capturing Global Structural Information in Long Document Question Answering with Compressive Graph Selector Network
  • DRLK: Dynamic Hierarchical Reasoning with Language Model and Knowledge Graph for Question Answering
  • ConvFinQA: Exploring the Chain of Numerical Reasoning in Conversational Finance Question Answering
  • "Video Question Answering: Datasets, Algorithms and Challenges"
  • Teaching Broad Reasoning Skills for Multi-Step QA by Generating Hard Contexts
  • PACIFIC: Towards Proactive Conversational Question Answering over Tabular and Textual Data in Finance
  • RLET: A Reinforcement Learning Based Approach for Explainable QA with Entailment Trees
  • monoQA: Multi-Task Learning of Reranking and Answer Extraction for Open-Retrieval Conversational Question Answering
  • UniRPG: Unified Discrete Reasoning over Table and Text as Program Generation
  • DuQM: A Chinese Dataset of Linguistically Perturbed Natural Questions for Evaluating the Robustness of Question Matching Models
  • TIARA: Multi-grained Retrieval for Robust Question Answering over Large Knowledge Base
  • Structure-Unified M-Tree Coding Solver for Math Word Problem
  • Rethinking Multi-Modal Alignment in Multi-Choice VideoQA from Feature and Sample Perspectives
  • ASQA: Factoid Questions Meet Long-Form Answers
  • Uni-Parser: Unified Semantic Parser for Question Answering on Knowledge Base and Database
  • ReasTAP: Injecting Table Reasoning Skills During Pre-training via Synthetic Reasoning Examples
  • Analogical Math Word Problems Solving with Enhanced Problem-Solution Association
  • Knowledge Transfer from Answer Ranking to Answer Generation
  • Empowering Language Models with Knowledge Graph Reasoning for Open-Domain Question Answering

19. Resources and Evaluation【资源和评测】 

  • On the Limitations of Reference-Free Evaluations of Generated Text
  • GuoFeng: A Benchmark for Zero Pronoun Recovery and Translation
  • GENIE: Toward Reproducible and Standardized Human Evaluation for Text Generation
  • Three Real-World Datasets and Neural Computational Models for Classification Tasks in Patent Landscaping
  • Agent-Specific Deontic Modality Detection in Legal Language
  • SCROLLS: Standardized CompaRison Over Long Language Sequences
  • "JDDC 2.1: A Multimodal Chinese Dialogue Dataset with Joint Tasks of Query Rewriting, Response Generation, Discourse Parsing, and  Summarization"
  • Multi-VQG: Generating Engaging Questions for Multiple Images
  • "Tomayto, Tomahto. Beyond Token-level Answer Equivalence for Question Answering Evaluation"
  • QRelScore: Better Evaluating Generated Questions with Deeper Understanding of Context-aware Relevance
  • Generative Language Models for Paragraph-Level Question Generation
  • "Cross-document Event Coreference Search: Task, Dataset and Modeling"
  • M2D2: A Massively Multi-Domain Language Modeling Dataset
  • "StoryER: Automatic Story Evaluation via Ranking, Rating and Reasoning"
  • Linguistic Corpus Annotation for Automatic Text Simplification Evaluation
  • Near-Negative Distinction: Giving a Second Life to Human Evaluation Datasets
  • Stanceosaurus: Classifying Stance Towards Multicultural Misinformation
  • When FLUE Meets FLANG: Benchmarks and Large Pretrained Language Model for Financial Domain
  • Reproducibility in Computational Linguistics: Is Source Code Enough?
  • MUSIED: A Benchmark for Event Detection from Multi-Source Heterogeneous Informal Texts
  • Reproducibility Issues for BERT-based Evaluation Metrics
  • On the Evaluation Metrics for Paraphrase Generation
  • A Second Wave of UD Hebrew Treebanking and Cross-Domain Parsing
  • MasakhaNER 2.0: Africa-centric Transfer Learning for Named Entity Recognition
  • ExPUNations: Augmenting Puns with Keywords and Explanations
  • Context Matters for Image Descriptions for Accessibility: Challenges for Referenceless Evaluation Metrics
  • MetaLogic: Logical Reasoning Explanations with Fine-Grained Structure
  • Super-NaturalInstructions: Generalization via Declarative Instructions on 1600+ NLP Tasks
  • RuCoLA: Russian Corpus of Linguistic Acceptability
  • Towards Knowledge-Intensive Text-to-SQL Semantic Parsing with Formulaic Knowledge
  • PHEE: A Dataset for Pharmacovigilance Event Extraction from Text
  • LILA: A Unified Benchmark for Mathematical Reasoning
  • Transfer Learning with Synthetic Corpora for Spatial Role Labeling and Reasoning
  • CEFR-Based Sentence Difficulty Annotation and Assessment
  • Title2Event: Benchmarking Open Event Extraction with a Large-scale Chinese Title Dataset
  • IDK-MRC: Unanswerable Questions for Indonesian Machine Reading Comprehension
  • FigMemes: A Dataset for Figurative Language Identification in Politically-Opinionated Memes
  • "ParaTag: A Dataset of Paraphrase Tagging for Fine-Grained Labels, NLG Evaluation, and Data Augmentation"
  • Open-domain Video Commentary Generation
  • EUR-Lex-Sum: A Multi- and Cross-lingual Dataset for Long-form Summarization in the Legal Domain
  • DivEMT: Neural Machine Translation Post-Editing Effort Across Typologically Diverse Languages
  • Human-Machine Collaboration Approaches to Build a Dialogue Dataset for Hate Speech Countering
  • Revisiting DocRED - Addressing the False Negative Problem in Relation Extraction
  • Bloom Library: Multimodal Datasets in 300+ Languages for a Variety of Downstream Tasks
  • Hierarchical Multi-Label Classification of Scientific Documents
  • Detecting Label Errors by Using Pre-Trained Language Models
  • CN-AutoMIC: Distilling Chinese Commonsense Knowledge from Pretrained Language Models
  • Improving Large-scale Paraphrase Acquisition and Generation
  • A Survey of Computational Framing Analysis Approaches
  • arXivEdits: Understanding the Human Revision Process in Scientific Writing
  • 🌾 DEMETR: Diagnosing Evaluation Metrics for Translation
  • RobustLR: A Diagnostic Benchmark for Evaluating Logical Robustness of Deductive Reasoners
  • CRIPP-VQA: Counterfactual Reasoning about Implicit Physical Properties via Video Question Answering
  • Exploring Document-Level Literary Machine Translation with Parallel Paragraphs from World Literature
  • A Dataset for Hyper-Relational Extraction and a Cube-Filling Approach
  • A Fine-grained Chinese Software Privacy Policy Dataset for Sequence Labeling and Regulation Compliant Identification
  • DiscoSense: Commonsense Reasoning with Discourse Connectives
  • Evaluating the Knowledge Dependency of Questions
  • Making Science Simple: Corpora for the Lay Summarisation of Scientific Literature
  • KOLD: Korean Offensive Language Dataset
  • ECTSum: A New Benchmark Dataset For Bullet Point Summarization of Long Earnings Call Transcripts

20. Semantics: Lexical, Sentence level, Textual Inference and Other areas【语义学:单词、句子层次、文本推理和其他领域】 

  • Measuring Context-Word Biases in Lexical Semantic Datasets
  • Unobserved Local Structures Make Compositional Generalization Hard
  • Mitigating Spurious Correlation in Natural Language Understanding with Counterfactual Inference
  • Mutual Exclusivity Training and Primitive Augmentation to Induce Compositionality
  • PCL: Peer-Contrastive Learning with Diverse Augmentations for Unsupervised Sentence Embeddings
  • UnifiedSKG: Unifying and Multi-Tasking Structured Knowledge Grounding with Text-to-Text Language Models
  • Reasoning Like Program Executors
  • DocInfer: Document-level Natural Language Inference using Optimal Evidence Selection
  • Infinite SCAN: An Infinite Model of Diachronic Semantic Change
  • "R2F: A General Retrieval, Reading and Fusion Framework for Document-level Natural Language Inference"
  • RASAT: Integrating Relational Structures into Pretrained Seq2Seq Model for Text-to-SQL
  • Sentence Representation Learning with Generative Objective rather than Contrastive Objective
  • Generating Literal and Implied Subquestions to Fact-check Complex Claims
  • Exploring the Secrets Behind the Learning Difficulty of Meaning Representations for Semantic Parsing
  • Understanding Jargon: Combining Extraction and Generation for Definition Modeling
  • Exploiting Global and Local Hierarchies for Hierarchical Text Classification
  • Semantic-aware Contrastive Learning for More Accurate Semantic Parsing
  • Kernel-Whitening: Overcome Dataset Bias with Isotropic Sentence Embedding
  • Inductive Relation Prediction with Logical Reasoning Using Contrastive Representations
  • Open World Classification with Adaptive Negative Samples
  • Neural-Symbolic Inference for Robust Autoregressive Graph Parsing via Compositional Uncertainty Quantification
  • Learning Semantic Textual Similarity via Topic-informed Discrete Latent Variables
  • Leveraging Affirmative Interpretations from Negation Improves Natural Language Understanding
  • GraphQ IR: Unifying the Semantic Parsing of Graph Query Languages with One Intermediate Representation
  • Retrofitting Multilingual Sentence Embeddings with Abstract Meaning Representation
  • DEER: Descriptive Knowledge Graph for Explaining Entity Relationships
  • FLUTE: Figurative Language Understanding through Textual Explanations
  • QASem Parsing: Text-to-text Modeling of QA-based Semantics
  • "Generate, Discriminate and Contrast: A Semi-Supervised Sentence Representation Learning Framework"
  • Natural Language Deduction with Incomplete Information
  • PromptBERT: Improving BERT Sentence Embeddings with Prompts
  • Are representations built from the ground up? An empirical examination of local composition in language models
  • Evaluating the Impact of Model Scale for Compositional Generalization in Semantic Parsing
  • Breakpoint Transformers for Modeling and Tracking Intermediate Beliefs
  • Looking at the Overlooked: An Analysis on the Word-Overlap Bias in Natural Language Inference
  • Cross-domain Generalization for AMR Parsing

21. Sentiment Analysis, Stylistic Analysis, and Argument Mining【情感分析、文体分析和论点挖掘】 

  • Semantic Simplification for Sentiment Classification
  • Curriculum Knowledge Distillation for Emoji-supervised Cross-lingual Sentiment Analysis
  • "A Multifaceted Framework to Evaluate Evasion, Content Preservation, and Misattribution in Authorship Obfuscation Techniques"
  • Mitigating Inconsistencies in Multimodal Sentiment Analysis under Uncertain Missing Modalities
  • Curriculum Learning Meets Weakly Supervised Multimodal Correlation Learning
  • COM-MRC: A COntext-Masked Machine Reading Comprehension Framework for Aspect Sentiment Triplet Extraction
  • CEM: Machine-Human Chatting Handoff via Causal-Enhance Module
  • Face-Sensitive Image-to-Emotional-Text Cross-modal Translation for Multimodal Aspect-based Sentiment Analysis
  • A Span-level Bidirectional Network for Aspect Sentiment Triplet Extraction
  • Efficient Nearest Neighbor Emotion Classification with BERT-whitening
  • Sentiment-Aware Word and Sentence Level Pre-training for Sentiment Analysis
  • AEG: Argumentative Essay Generation via A Dual-Decoder Model with Content Planning
  • Pair-Based Joint Encoding with Relational Graph Convolutional Networks for Emotion-Cause Pair Extraction
  • AX-MABSA: A Framework for Extremely Weakly Supervised Multi-label Aspect Based Sentiment Analysis
  • Generative Data Augmentation with Contrastive Learning for Zero-Shot Stance Detection
  • Text Style Transferring via Adversarial Masking and Styled Filling
  • UniMSE: Towards Unified Multimodal Sentiment Analysis and Emotion Recognition
  • Improving Aspect Sentiment Quad Prediction via Template-Order Data Augmentation
  • Generative Entity-to-Entity Stance Detection with Knowledge Graph Augmentation
  • Symptom Identification for Interpretable Detection of Multiple Mental Disorders on Social Media
  • A Simple Contrastive Learning Framework for Interactive Argument Pair Identification via Argument-Context Extraction
  • Prompt-based Distribution Alignment for Domain Generalization in Text Classification
  • A Generative Model for End-to-End Argument Mining with Reconstructed Positional Encoding and Constrained Pointer Mechanism

22. Speech, Vision, Robotics, Multimodal Grounding【语音、视觉、机器人、多模态】 

  • PEVL: Position-enhanced Pre-training and Prompt Tuning for Vision-language Models
  • Textless Speech Emotion Conversion using Discrete & Decomposed Representations
  • Retrieval Augmented Visual Question Answering with Outside Knowledge
  • Normalized Contrastive Learning for Text-Video Retrieval
  • Robustness of Fusion-based Multimodal Classifiers to Cross-Modal Content Dilutions
  • Abstract Visual Reasoning with Tangram Shapes
  • Z-LaVI: Zero-Shot Language Solver Fueled by Visual Imagination
  • DANLI: Deliberative Agent for Following Natural Language Instructions
  • SpeechUT: Bridging Speech and Text with Hidden-Unit for Encoder-Decoder Based Speech-Text Pre-training
  • Can Visual Context Improve Automatic Speech Recognition for an Embodied Agent?
  • Why is Winoground Hard? Investigating Failures in Visuolinguistic Compositionality
  • LVP-M3: Language-aware Visual Prompt for Multilingual Multimodal Machine Translation
  • UniGeo: Unifying Geometry Logical Reasoning via Reformulating Mathematical Expression
  • CPL: Counterfactual Prompt Learning for Vision and Language Models
  • MGDoc: Pre-training with Multi-granular Hierarchy for Document Image Understanding
  • TRIPS: Efficient Vision-and-Language Pre-training with Text-Relevant Image Patch Selection
  • RelCLIP: Adapting Language-Image Pretraining for Visual Relationship Detection via Relational Contrastive Learning
  • McQueen: a Benchmark for Multimodal Conversational Query Rewrite
  • Discrete Cross-Modal Alignment Enables Zero-Shot Speech Translation
  • An Anchor-based Relative Position Embedding Method for Cross-Modal Tasks
  • GHAN: Graph-Based Hierarchical Aggregation Network for Text-Video Retrieval
  • A Span-based Multimodal Variational Autoencoder for Semi-supervised Multimodal Named Entity Recognition
  • Open-Domain Sign Language Translation Learned from Online Video
  • ULN: Towards Underspecified Vision-and-Language Navigation
  • Towards Unifying Reference Expression Generation and Comprehension
  • mPLUG: Effective and Efficient Vision-Language Learning by Cross-modal Skip-connections
  • Speaker Overlap-aware Neural Diarization for Multi-party Meeting Analysis
  • SEMGraph: Incorporating Sentiment Knowledge and Eye Movement into Graph Model for Sentiment Analysis
  • Extending Phrase Grounding with Pronouns in Visual Dialogues
  • LiteVL: Efficient Video-Language Learning with Enhanced Spatial-Temporal Modeling
  • Entity-Focused Dense Passage Retrieval for Outside-Knowledge Visual Question Answering
  • Character-centric Story Visualization via Visual Planning and Token Alignment
  • Contrastive Learning with Expectation-Maximization for Weakly Supervised Phrase Grounding
  • Distilled Dual-Encoder Model for Vision-Language Understanding
  • Weakly-Supervised Temporal Article Grounding
  • Adaptive Contrastive Learning on Multimodal Transformer for Review Helpfulness Prediction
  • FaD-VLP: Fashion Vision-and-Language Pre-training towards Unified Retrieval and Captioning
  • End-to-End Unsupervised Vision-and-Language Pre-training with Referring Expression Matching

23. Summarization【文本摘要】 

  • How Far are We from Robust Long Abstractive Summarization?
  • CiteSum: Citation Text-guided Scientific Extreme Summarization and Domain Adaptation with Limited Supervision
  • Learning to Generate Overlap Summaries through Noisy Synthetic Data
  • Toward Unifying Text Segmentation and Long Document Summarization
  • SNaC: Coherence Error Detection for Narrative Summarization
  • HydraSum: Disentangling Style Features in Text Summarization with Multi-Decoder Models
  • SEM-F1: an Automatic Way for Semantic Evaluation of Multi-Narrative Overlap Summaries at Scale
  • SQuALITY: Building a Long-Document Summarization Dataset the Hard Way
  • Effective and Efficient Query-aware Snippet Extraction for Web Search
  • Opinion Summarization by Weak-Supervision from Mix-structured Data
  • Few-shot Query-Focused Summarization with Prefix-Merging
  • Summarizing Community-based Question-Answer Pairs
  • Scientific Paper Extractive Summarization Enhanced by Citation Graphs
  • Analyzing and Evaluating Faithfulness in Dialogue Summarization
  • Abstractive Summarization Guided by Latent Hierarchical Document Structure
  • RACE: Retrieval-augmented Commit Message Generation
  • CTRLsum: Towards Generic Controllable Text Summarization
  • Leveraging Locality in Abstractive Text Summarization
  • Salience Allocation as Guidance for Abstractive Summarization
  • Factorizing Content and Budget Decisions in Abstractive Summarization of Long Documents
  • Assist Non-native Viewers: Multimodal Cross-Lingual Summarization for How2 Videos
  • X-FACTOR: A Cross-metric Evaluation of Factual Correctness in Abstractive Summarization
  • ClidSum: A Benchmark Dataset for Cross-Lingual Dialogue Summarization
  • Towards Summary Candidates Fusion
  • Unsupervised Opinion Summarisation in the Wasserstein Space
  • Why Do You Feel This Way? Summarizing Triggers of Emotions in Social Media Posts
  • Evaluating and Improving Factuality in Multimodal Abstractive Summarization
  • Referee: Reference-Free Sentence Summarization with Sharper Controllability through Symbolic Knowledge Distillation
  • Learning with Rejection for Abstractive Text Summarization
  • Correcting Diverse Factual Errors in Abstractive Summarization via Post-Editing and Language Model Infilling
  • HEGEL: Hypergraph Transformer for Long Document Summarization


24. Syntax, Parsing and their Applications【语法、句法及其应用】 

  • SynGEC: Syntax-Enhanced Grammatical Error Correction with a Tailored GEC-Oriented Parser
  • Learning a Grammar Inducer from Massive Uncurated Instructional Videos
  • Unbiased and Efficient Sampling of Dependency Trees
  • Algorithms for Acyclic Weighted Finite-State Automata with Failure Arcs
  • On Parsing as Tagging
  • Algorithms for Weighted Pushdown Automata
  • "Reorder and then Parse, Fast and Accurate Discontinuous Constituency Parsing"

25. Theme Track【主题追踪】 

  • Towards Climate Awareness in NLP Research
  • Whose Language Counts as High Quality? Measuring Language Ideologies in Text Data Selection
  • Geographic Citation Gaps in NLP Research
  • Neural Theory-of-Mind? On the Limits of Social Intelligence in Large LMs
  • Structural generalization is hard for sequence-to-sequence models
  • Questioning the Validity of Summarization Datasets and Improving Their Factual Consistency
  • Missing Counter-Evidence Renders NLP Fact-Checking Unrealistic for Misinformation
  • The Authenticity Gap in Human Evaluation
  • Counterfactual Recipe Generation: Exploring Compositional Generalization in a Realistic Scenario
  • Bridging Fairness and Environmental Sustainability in Natural Language Processing
  • Towards Robust Numerical Question Answering: Diagnosing Numerical Capabilities of NLP Systems
  • CONDAQA: A Contrastive Reading Comprehension Dataset for Reasoning about Negation

26. Unsupervised and Weakly-Supervised Methods in NLP【自然语言处理中的无监督和弱监督方法】


  • Bilingual Lexicon Induction for Low-Resource Languages using Graph Matching via Optimal Transport

  • CycleKQR: Unsupervised Bidirectional Keyword-Question Rewriting

  • Zero-Shot Text Classification with Self-Training

  • Fine-grained Category Discovery under Coarse-grained supervision with Hierarchical Weighted Self-contrastive Learning

  • Learning Instructions with Unlabeled Data for Zero-Shot Cross-Task Generalization

  • Fast-R2D2: A Pretrained Recursive Neural Network based on Pruned CKY for Grammar Induction and Text Representation

  • Unsupervised Tokenization Learning

  • FastClass: A Time-Efficient Approach to Weakly-Supervised Text Classification

  • PASTA: Table-Operations Aware Fact Verification via Sentence-Table Cloze Pre-training

  • Rethinking Style Transformer with Energy-based Interpretation: Adversarial Unsupervised Style Transfer using a Pretrained Model

  • IsoVec: Controlling the Relative Isomorphism of Word Embedding Spaces

  • Modal-specific Pseudo Query Generation for Video Corpus Moment Retrieval

  • Beyond prompting: Making Pre-trained Language Models Better Zero-shot Learners by Clustering Representations


更多推荐



Fast and Effective!一文速览轻量化Transformer各领域研究进展

ICLR 2023(投稿)|自然语言处理相关论文分类整理

CIKM 2022 | 推荐系统相关论文分类整理



点击下方“阅读原文”前往知乎专栏
↓↓↓

您可能也对以下帖子感兴趣

文章有问题?点此查看未经处理的缓存