EMNLP 2022 Accepted Papers
Motivation
EMNLP 2022 is as great as can be, but navigating the official lists of accepted papers proves tricky. Here’s a practical, mobile-friendly, list of all accepted papers at EMNLP 2022. Click on sessions to expend/reduce them.
Note that the official website provides two documents:
- an excel file with titles, authors, tracks
- a PDF with timestamps and abstracts
This page has been (mostly) automatically generated using both. If you spot a mistake, I’m happy to fix it, just send me an email! Same goes if you have a clever idea on how to add links to papers.
Friday 9th December
11:00-11:15
RankGen: Improving Text Generation with Large Ranking Models
11:15-11:30
Linearizing Transformer with Key-Value Memory
11:30-11:45
A Unified Encoder-Decoder Framework with Entity Memory
11:45-12:00
A Distributional Lens for Multi-Aspect Controllable Text Generation
12:00-12:15
ELMER: A Non-Autoregressive Pre-trained Language Model for Efficient and Effective Text Generation
12:15-12:30
Curriculum Prompt Learning with Self-Training for Abstractive Dialogue Summarization
11:00-11:15
Multi-VQG: Generating Engaging Questions for Multiple Images
11:15-11:30
Tomayto, Tomahto. Beyond Token-level Answer Equivalence for Question Answering Evaluation
11:30-11:45
QRelScore: Better Evaluating Generated Questions with Deeper Understanding of Context-aware Relevance
11:45-12:00
Generative Language Models for Paragraph-Level Question Generation
12:00-12:15
Cross-document Event Coreference Search: Task, Dataset and Modeling
12:15-12:30
M2D2: A Massively Multi-Domain Language Modeling Dataset
11:00-11:15
UnifiedSKG: Unifying and Multi-Tasking Structured Knowledge Grounding with Text-to-Text Language Models
11:15-11:30
Reasoning Like Program Executors
11:30-11:45
DocInfer: Document-level Natural Language Inference using Optimal Evidence Selection
11:45-12:00
Infinite SCAN: An Infinite Model of Diachronic Semantic Change
12:00-12:15
Measuring Context-Word Biases in Lexical Semantic Datasets
12:15-12:30
Unobserved Local Structures Make Compositional Generalization Hard
11:00-11:15
Toward Unifying Text Segmentation and Long Document Summarization
11:15-11:30
SNaC: Coherence Error Detection for Narrative Summarization
11:30-11:45
HydraSum: Disentangling Style Features in Text Summarization with Multi-Decoder Models
11:45-12:00
SEM-F1: an Automatic Way for Semantic Evaluation of Multi-Narrative Overlap Summaries at Scale
12:00-12:15
SQuALITY: Building a Long-Document Summarization Dataset the Hard Way
12:15-12:30
How Far are We from Robust Long Abstractive Summarization?
11:00-11:15
[INDUSTRY] Improving Large-Scale Conversational Assistants using Model Interpretation based Training Sample Selection
11:15-11:30
[INDUSTRY] CGF: Constrained Generation Framework for Query Rewriting in Conversational AI
11:30-11:45
[INDUSTRY] SimANS: Simple Ambiguous Negatives Sampling for Dense Text Retrieval
11:45-12:00
[INDUSTRY] Learning Geolocations for Cold-Start and Hard-to-Resolve Addresses via Deep Metric Learning
12:00-12:15
[INDUSTRY] Large-scale Machine Translation for Indian Languages in E-commerce under Low Resource Constraints
12:15-12:30
[INDUSTRY] Improving Text-to-SQL Semantic Parsing with Fine-grained Query Understanding
11:00-11:15
Learning to Generate Question by Asking Question: A Primal-Dual Approach with Uncommon Word Generation
11:15-11:30
PAIR: Prompt-Aware margIn Ranking for Counselor Reflection Scoring in Motivational Interviewing
11:30-11:45
Robustness of Fusion-based Multimodal Classifiers to Cross-Modal Content Dilutions
11:45-12:00
Translation between Molecules and Natural Language
12:00-12:15
Guiding Neural Entity Alignment with Compatibility
12:15-12:30
How Large Language Models are Transforming Machine-Paraphrase Plagiarism
11:00-12:30
TranSHER: Translating Knowledge Graph Embedding with Hyper-Ellipsoidal Restriction
11:00-12:30
Robots-Dont-Cry: Understanding Falsely Anthropomorphic Utterances in Dialog Systems
11:00-12:30
When More Data Hurts: A Troubling Quirk in Developing Broad-Coverage Natural Language Understanding Systems
11:00-12:30
Less is More: Summary of Long Instructions is Better for Program Synthesis
11:00-12:30
HashFormers: Towards Vocabulary-independent Pre-trained Transformers
11:00-12:30
AMAL: Meta Knowledge-Driven Few-Shot Adapter Learning
11:00-12:30
Facilitating Contrastive Learning of Discourse Relational Senses by Exploiting the Hierarchy of Sense Relations
11:00-12:30
Quantifying Privacy Risks of Masked Language Models Using Membership Inference Attacks
11:00-12:30
MatchPrompt: Prompt-based Open Relation Extraction with Semantic Consistency Guided Clustering
11:00-12:30
Incorporating Relevance Feedback for Information-Seeking Retrieval using Few-Shot Document Re-Ranking
11:00-12:30
Extending Logic Explained Networks to Text Classification
11:00-12:30
Are All Spurious Features in Natural Language Alike? An Analysis through a Causal Lens
11:00-12:30
Human Guided Exploitation of Interpretable Attention Patterns in Summarization and Topic Segmentation
11:00-12:30
InforMask: Unsupervised Informative Masking for Language Model Pretraining
11:00-12:30
Subword Evenness (SuE) as a Predictor of Cross-lingual Transfer to Low-resource Languages
11:00-12:30
Don’t Prompt, Search! Mining-based Zero-Shot Learning with Language Models
11:00-12:30
Fine-Tuning Pre-trained Transformers into Decaying Fast Weights
11:00-12:30
Efficient Large Scale Language Modeling with Mixtures of Experts
11:00-12:30
The Curious Case of Control
11:00-12:30
RLPrompt: Optimizing Discrete Text Prompts with Reinforcement Learning
11:00-12:30
Natural Language to Code Translation with Execution
11:00-12:30
[CL] Neural Embedding Allocation: Distributed Representations of Topic Models
11:00-12:30
Chunk-based Nearest Neighbor Machine Translation
11:00-12:30
ConNER: Consistency Training for Cross-lingual Named Entity Recognition
11:00-12:30
Transforming Sequence Tagging Into A Seq2Seq Task
11:00-12:30
[INDUSTRY] Accelerating the Discovery of Semantic Associations from Medical Literature: Mining Relations Between Diseases and Symptoms
11:00-12:30
[INDUSTRY] Machine translation impact in E-commerce multilingual search
11:00-12:30
[INDUSTRY] Exploiting In-Domain Bilingual Corpora for Zero-Shot Transfer Learning in NLU of Intra-Sentential Code-Switching Chatbot Interactions
11:00-12:30
[INDUSTRY] Calibrating Imbalanced Classifiers with Focal Loss: An Empirical Study
11:00-12:30
[INDUSTRY] Unsupervised training data re-weighting for natural language understanding with local distribution approximation
11:00-12:30
[INDUSTRY] Cross-Encoder Data Annotation for Bi-Encoder Based Product Matching
11:00-12:30
[INDUSTRY] Multi-Tenant Optimization For Few-Shot Task-Oriented FAQ Retrieval
11:00-12:30
Life is a Circus and We are the Clowns: Automatically Finding Analogies between Situations and Processes
11:00-12:30
Automatic Generation of Socratic Subquestions for Teaching Math Word Problems
11:00-12:30
Factual Accuracy is not Enough: Planning Consistent Description Order for Radiology Report Generation
11:00-12:30
Differentially Private Language Models for Secure Data Sharing
11:00-12:30
Hard Gate Knowledge Distillation - Leverage Calibration for Robust and Reliable Language Model
11:00-12:30
Improving Iterative Text Revision by Learning Where to Edit from Other Revision Tasks
11:00-12:30
FiE: Building a Global Probability Space by Leveraging Early Fusion in Encoder for Open-Domain Question Answering
11:00-12:30
monoQA: Multi-Task Learning of Reranking and Answer Extraction for Open-Retrieval Conversational Question Answering
11:00-12:30
Empowering Language Models with Knowledge Graph Reasoning for Open-Domain Question Answering
11:00-12:30
FigMemes: A Dataset for Figurative Language Identification in Politically-Opinionated Memes
11:00-12:30
Detecting Label Errors by Using Pre-Trained Language Models
11:00-12:30
Evaluating the Knowledge Dependency of Questions
11:00-12:30
On the Limitations of Reference-Free Evaluations of Generated Text
11:00-12:30
Three Real-World Datasets and Neural Computational Models for Classification Tasks in Patent Landscaping
11:00-12:30
Natural Language Deduction with Incomplete Information
11:00-12:30
Evaluating the Impact of Model Scale for Compositional Generalization in Semantic Parsing
11:00-12:30
Mitigating Spurious Correlation in Natural Language Understanding with Counterfactual Inference
11:00-12:30
TRIPS: Efficient Vision-and-Language Pre-training with Text-Relevant Image Patch Selection
11:00-12:30
Adaptive Contrastive Learning on Multimodal Transformer for Review Helpfulness Prediction
11:00-12:30
Mutual Information Alleviates Hallucinations in Abstractive Summarization
11:00-12:30
Salience Allocation as Guidance for Abstractive Summarization
11:00-12:30
Improving Factual Consistency in Summarization with Compression-Based Post-Editing
11:00-12:30
Learning with Rejection for Abstractive Text Summarization
11:00-12:30
Fast-R2D2: A Pretrained Recursive Neural Network based on Pruned CKY for Grammar Induction and Text Representation
11:00-12:30
[DEMO] LM-Debugger: An Interactive Tool for Inspection and Intervention in Transformer-Based Language Models
11:00-12:30
[DEMO] FairLib: A Unified Framework for Assessing and Improving Fairness
11:00-12:30
[DEMO] Snoopy: An Online Interface for Exploring the Effect of Pretraining Term Frequencies on Few-Shot LM Performance
11:00-12:30
[DEMO] Azimuth: Systematic Error Analysis for Text Classification
14:00-14:15
The Geometry of Multilingual Language Model Representations
14:15-14:30
What Makes Instruction Learning Hard? An Investigation and a New Challenge in a Synthetic Environment
14:30-14:45
Language Model Pre-Training with Sparse Latent Typing
14:45-15:00
Ground-Truth Labels Matter: A Deeper Look into Input-Label Demonstrations
15:00-15:15
Language Model Decomposition: Quantifying the Dependency and Correlation of Language Models
15:15-15:30
Iteratively Prompt Pre-trained Language Models for Chain of Thought
14:00-14:15
Curriculum Knowledge Distillation for Emoji-supervised Cross-lingual Sentiment Analysis
14:15-14:30
Sentence-Incremental Neural Coreference Resolution
14:30-14:45
A Multifaceted Framework to Evaluate Evasion, Content Preservation, and Misattribution in Authorship Obfuscation Techniques
14:45-15:00
Affective Idiosyncratic Responses to Music
15:00-15:15
Varifocal Question Generation for Fact-checking
15:15-15:30
Topic-Regularized Authorship Representation Learning
14:00-14:15
Normalized Contrastive Learning for Text-Video Retrieval
14:15-14:30
Abstract Visual Reasoning with Tangram Shapes
14:30-14:45
Z-LaVI: Zero-Shot Language Solver Fueled by Visual Imagination
14:45-15:00
DANLI: Deliberative Agent for Following Natural Language Instructions
15:00-15:15
Learning a Grammar Inducer from Massive Uncurated Instructional Videos
15:15-15:30
How well can Text-to-Image Generative Models understand Ethical Natural Language Interventions?
14:00-14:15
Generating Natural Language Proofs with Verifier-Guided Search
14:15-14:30
Improving Complex Knowledge Base Question Answering via Question-to-Action and Question-to-Question Alignment
14:30-14:45
Successive Prompting for Decomposing Complex Questions
14:45-15:00
M3: A Multi-View Fusion and Multi-Decoding Network for Multi-Document Reading Comprehension
15:00-15:15
Semantic Framework based Query Generation for Temporal Question Answering over Knowledge Graphs
15:15-15:30
Improving compositional generalization for multi-step quantitative reasoning in question answering
14:00-14:15
[TACL] OPAL: Ontology-Aware Pretrained Language Model for End-to-End Task-Oriented Dialogue
14:15-14:30
[TACL] True Few-Shot Learning With Prompts - A Real-World Perspective
14:30-14:45
[TACL] Generate, Annotate, and Learn: NLP with Synthetic Text
14:45-15:00
[TACL] Structural Persistence in Language Models: Priming as a Window into Abstract Language Representations
15:00-15:15
[TACL] ProoFVer: Natural Logic Theorem Proving for Fact Verification
15:15-15:30
[TACL] Investigating Reasons for Disagreement in Natural Language Inference
14:00-14:15
Gendered Mental Health Stigma in Masked Language Models
14:15-14:30
SafeText: A Benchmark for Exploring Physical Safety in Language Models
14:30-14:45
Prompting for Multimodal Hateful Meme Classification
14:45-15:00
Modeling Information Change in Science Communication with Semantically Matched Paraphrases
15:00-15:15
Tracing Semantic Variation in Slang
15:15-15:30
An Empirical Analysis of Memorization in Fine-tuned Autoregressive Language Models
16:00-17:30
ReCo: Reliable Causal Chain Reasoning via Structural Causal Recurrent Neural Networks
16:00-17:30
A Sequential Flow Control Framework for Multi-hop Knowledge Base Question Answering
16:00-17:30
Identifying Physical Object Use in Sentences
16:00-17:30
Sequence Models for Document Structure Identification in an Undeciphered Script
16:00-17:30
Sentence-level Media Bias Analysis Informed by Discourse Structures
16:00-17:30
META-GUI: Towards Multi-modal Conversational Agents on Mobile GUI
16:00-17:30
UniNL: Aligning Representation Learning with Scoring Function for OOD Detection via Unified Neighborhood Learning
16:00-17:30
Prompt Conditioned VAE: Enhancing Generative Replay for Lifelong Learning in Task-Oriented Dialogue
16:00-17:30
End-to-End Neural Discourse Deixis Resolution in Dialogue
16:00-17:30
Sparse Teachers Can Be Dense with Knowledge
16:00-17:30
Vector-Quantized Input-Contextualized Soft Prompts for Natural Language Understanding
16:00-17:30
"I’m sorry to hear that": Finding New Biases in Language Models with a Holistic Descriptor Dataset
16:00-17:30
BERTScore is Unfair: On Social Bias in Language Model-Based Metrics for Text Generation
16:00-17:30
TextFusion: Privacy-Preserving Pre-trained Model Inference via Token Fusion
16:00-17:30
[DEMO] Twitter-Demographer: A Flow-based Tool to Enrich Twitter Data
16:00-17:30
[DEMO] Hands-On Interactive Neuro-Symbolic NLP with DRaiL
16:00-17:30
Bi-Directional Iterative Prompt-Tuning for Event Argument Extraction
16:00-17:30
Attention and Edge-Label Guided Graph Convolutional Networks for Named Entity Recognition
16:00-17:30
Open Relation and Event Type Discovery with Type Abstraction
16:00-17:30
WR-One2Set: Towards Well-Calibrated Keyphrase Generation
16:00-17:30
OTSeq2Set: An Optimal Transport Enhanced Sequence-to-Set Model for Extreme Multi-label Text Classification
16:00-17:30
Dimension Reduction for Efficient Dense Retrieval via Conditional Autoencoder
16:00-17:30
A Framework for Adapting Pre-Trained Language Models to Knowledge Graph Completion
16:00-17:30
A Unified Neural Network Model for Readability Assessment with Feature Projection and Length-Balanced Loss
16:00-17:30
Recovering Gold from Black Sand: Multilingual Dense Passage Retrieval with Hard and False Negative Samples
16:00-17:30
Calibration Meets Explanation: A Simple and Effective Approach for Model Confidence Estimates
16:00-17:30
Towards Interactivity and Interpretability: A Rationale-based Legal Judgment Prediction Framework
16:00-17:30
TASA: Deceiving Question Answering Models by Twin Answer Sentences Attack
16:00-17:30
Improving Temporal Generalization of Pre-trained Language Models with Lexical Semantic Change
16:00-17:30
Exploring Mode Connectivity for Pre-trained Language Models
16:00-17:30
Parameter-Efficient Tuning Makes a Good Classification Head
16:00-17:30
[INDUSTRY] A Hybrid Approach to Cross-lingual Product Review Summarization
16:00-17:30
[INDUSTRY] Knowledge Distillation based Contextual Relevance Matching for E-commerce Product Search
16:00-17:30
[INDUSTRY] Tackling Temporal Questions in Natural Language Interface to Databases
16:00-17:30
[INDUSTRY] Unsupervised Dense Retrieval for Scientific Articles
16:00-17:30
[INDUSTRY] Developing Prefix-Tuning Models for Hierarchical Text Classification
16:00-17:30
Due to rapidly growing cyber-attacks and security vulnerabilities, many reports on cyber-threat intelligence are being published daily.
16:00-17:30
[INDUSTRY] Revisiting and Advancing Chinese Natural Language Understanding with Accelerated Heterogeneous Knowledge Pretraining
16:00-17:30
[INDUSTRY] PILE: Pairwise Iterative Logits Ensemble for Multi-Teacher Labeled Distillation
16:00-17:30
Active Example Selection for In-Context Learning
16:00-17:30
BBTv2: Towards a Gradient-Free Future with Large Language Models
16:00-17:30
G-MAP: General Memory-Augmented Pre-trained Language Model for Domain Tasks
16:00-17:30
Textual Manifold-based Defense Against Natural Language Adversarial Examples
16:00-17:30
[CL] Enhancing Lifelong Language Learning by Improving Pseudo-Sample Generation
16:00-17:30
Neural Machine Translation with Contrastive Translation Memories
16:00-17:30
A Template-based Method for Constrained Neural Machine Translation
16:00-17:30
Competency-Aware Neural Machine Translation: Can Machine Translation Know its Own Translation Quality?
16:00-17:30
Multi-Granularity Optimization for Non-Autoregressive Translation
16:00-17:30
Improving Machine Translation with Phrase Pair Injection and Corpus Filtering
16:00-17:30
XLM-D: Decorate Cross-lingual Pre-training Model as Non-Autoregressive Neural Machine Translation
16:00-17:30
ConsistTL: Modeling Consistency in Transfer Learning for Low-Resource Neural Machine Translation
16:00-17:30
RAPO: An Adaptive Ranking Paradigm for Bilingual Lexicon Induction
16:00-17:30
Entropy-Based Vocabulary Substitution for Incremental Learning in Multilingual Neural Machine Translation
16:00-17:30
Digging Errors in NMT: Evaluating and Understanding Model Errors from Partial Hypothesis Space
16:00-17:30
Cross-Align: Modeling Deep Cross-lingual Interactions for Word Alignment
16:00-17:30
Discovering Low-rank Subspaces for Language-agnostic Multilingual Representations
16:00-17:30
Intriguing Properties of Compression on Multilingual Models
16:00-17:30
English Contrastive Learning Can Learn Universal Cross-lingual Sentence Embeddings
16:00-17:30
PromptEHR: Conditional Electronic Healthcare Records Generation with Prompt Learning
16:00-17:30
Rethinking Positional Encoding in Tree Transformer for Code Representation
16:00-17:30
Chapter Ordering in Novels
16:00-17:30
Open-ended Knowledge Tracing for Computer Science Education
16:00-17:30
SEEN: Structured Event Enhancement Network for Explainable Need Detection of Information Recall Assistance
16:00-17:30
Tiny-NewsRec: Effective and Efficient PLM-based News Recommendation
16:00-17:30
Boundary-Driven Table-Filling for Aspect Sentiment Triplet Extraction
Computer-aided translation play a prominent role in the translation workflow of professional
Cross-lingual neural fuzzy matching for exploiting target-language monolingual corpora in computer-aided translation
16:00-17:30
Improved grammatical error correction by ranking elementary edits
16:00-17:30
Keyphrase Generation via Soft and Hard Semantic Corrections
16:00-17:30
JANUS: Joint Autoregressive and Non-autoregressive Training with Auxiliary Loss for Sequence Generation
16:00-17:30
MOCHA: A Multi-Task Training Approach for Coherent Text Generation from Cognitive Perspective
16:00-17:30
[DEMO] AGReE: A system for generating Automated Grammar Reading Exercises
16:00-17:30
[DEMO] Automatic Comment Generation for Chinese Student Narrative Essays
16:00-17:30
Eeny, meeny, miny, moe. How to choose data for morphological inflection.
16:00-17:30
Explainable Question Answering based on Semantic Graph by Global Differentiable Learning and Dynamic Adaptive Reasoning
16:00-17:30
Teaching Broad Reasoning Skills for Multi-Step QA by Generating Hard Contexts
16:00-17:30
DuQM: A Chinese Dataset of Linguistically Perturbed Natural Questions for Evaluating the Robustness of Question Matching Models
16:00-17:30
Structure-Unified M-Tree Coding Solver for Math Word Problem
16:00-17:30
Graph-Induced Transformers for Efficient Multi-Hop Question Answering
16:00-17:30
Pre-training Language Models with Deterministic Factual Knowledge
16:00-17:30
OpenCQA: Open-ended Question Answering with Charts
16:00-17:30
Title2Event: Benchmarking Open Event Extraction with a Large-scale Chinese Title Dataset
16:00-17:30
CN-AutoMIC: Distilling Chinese Commonsense Knowledge from Pretrained Language Models
16:00-17:30
Improving Large-scale Paraphrase Acquisition and Generation
16:00-17:30
A Survey of Computational Framing Analysis Approaches
16:00-17:30
CRIPP-VQA: Counterfactual Reasoning about Implicit Physical Properties via Video Question Answering
16:00-17:30
GuoFeng: A Benchmark for Zero Pronoun Recovery and Translation
16:00-17:30
Effective and Efficient Query-aware Snippet Extraction for Web Search
16:00-17:30
Opinion Summarization by Weak-Supervision from Mix-structured Data
16:00-17:30
Improving Faithfulness by Augmenting Negative Summaries from Fake Documents
16:00-17:30
[DEMO] MIC: A Multi-task Interactive Curation Tool
16:00-17:30
[DEMO] POTATO: The Portable Text Annotation Tool
16:00-17:30
Kernel-Whitening: Overcome Dataset Bias with Isotropic Sentence Embedding
16:00-17:30
Neural-Symbolic Inference for Robust Autoregressive Graph Parsing via Compositional Uncertainty Quantification
16:00-17:30
Leveraging Affirmative Interpretations from Negation Improves Natural Language Understanding
16:00-17:30
PromptBERT: Improving BERT Sentence Embeddings with Prompts
16:00-17:30
An Empirical Revisiting of Linguistic Knowledge Fusion in Language Understanding Tasks
16:00-17:30
Cross-domain Generalization for AMR Parsing
16:00-17:30
Mutual Exclusivity Training and Primitive Augmentation to Induce Compositionality
16:00-17:30
Mitigating Inconsistencies in Multimodal Sentiment Analysis under Uncertain Missing Modalities
16:00-17:30
Pair-Based Joint Encoding with Relational Graph Convolutional Networks for Emotion-Cause Pair Extraction
16:00-17:30
UniMSE: Towards Unified Multimodal Sentiment Analysis and Emotion Recognition
16:00-17:30
Argument Mining for Review Helpfulness Prediction
16:00-17:30
Prompt-based Distribution Alignment for Domain Generalization in Text Classification
16:00-17:30
A Generative Model for End-to-End Argument Mining with Reconstructed Positional Encoding and Constrained Pointer Mechanism
16:00-17:30
Semantic Simplification for Sentiment Classification
16:00-17:30
[CL] Dimensional Modeling of Emotions in Text with Appraisal Theories: Corpus Creation, Annotation Reliability, and Prediction
16:00-17:30
LiteVL: Efficient Video-Language Learning with Enhanced Spatial-Temporal Modeling
16:00-17:30
ALFRED-L: Investigating the Role of Language for Action Learning in Interactive Visual Environments
16:00-17:30
Directions for NLP Practices Applied to Online Hate Speech Detection
16:00-17:30
[DEMO] An Explainable Toolbox for Evaluating Pre-trained Vision-Language Models
16:00-17:30
Structural Constraints and Natural Language Inference for End-to-End Flowchart Grounded Dialog Response Generation
16:00-17:30
Should We Ban English NLP for a Year?
16:00-17:30
That’s the Wrong Lung! Evaluating and Improving the Interpretability of Unsupervised Multimodal Encoders for Medical Data
16:00-17:30
Adversarial Concept Erasure in Kernel Space
16:00-17:30
One size does not fit all: Investigating strategies for differentially-private learning across NLP tasks
16:00-17:30
Towards Teachable Reasoning Systems: Using a Dynamic Memory of User Feedback for Continual System Improvement
16:00-17:30
Mixed-effects transformers for hierarchical adaptation
16:00-17:30
Adapting a Language Model While Preserving its General Knowledge
16:00-17:30
Towards Opening the Black Box of Neural Machine Translation: Source and Target Interpretations of the Transformer
16:00-17:30
Polyglot Prompt: Multilingual Multitask Prompt Training
16:00-17:30
Context-Situated Pun Generation
16:00-17:30
Twist Decoding: Diverse Generators Guide Each Other
16:00-17:30
T-STAR: Truthful Style Transfer using AMR Graph as Intermediate Representation
16:00-17:30
LILA: A Unified Benchmark for Mathematical Reasoning
16:00-17:30
Character-centric Story Visualization via Visual Planning and Token Alignment
16:00-17:30
Algorithms for Acyclic Weighted Finite-State Automata with Failure Arcs
16:00-17:30
The "Problem” of Human Label Variation: On Ground Truth in Data, Modeling and Evaluation
Saturday 10th December
09:00-09:15
CDConv: A Benchmark for Contradiction Detection in Chinese Conversations
09:15-09:30
Co-guiding Net: Achieving Mutual Guidances between Multiple Intent Detection and Slot Filling via Heterogeneous Semantics-Label Graphs
09:30-09:45
Estimating Soft Labels for Out-of-Domain Intent Detection
09:45-10:00
InstructDial: Improving Zero and Few-shot Generalization in Dialogue through Instruction Tuning
10:00-10:15
Aligning Recommendation and Conversation via Dual Imitation
10:15-10:30
Correctable-DST: Mitigating Historical Context Mismatch between Training and Inference for Improved Dialogue State Tracking
09:00-09:15
Crossmodal-3600: A Massively Multilingual Multimodal Evaluation Dataset
09:15-09:30
Graph-Based Multilingual Label Propagation for Low-Resource Part-of-Speech Tagging
09:30-09:45
AfroLID: A Neural Language Identification Tool for African Languages
09:45-10:00
The (Undesired) Attenuation of Human Biases by Multilinguality
10:00-10:15
CoCoa: An Encoder-Decoder Model for Controllable Code-switched Generation
10:15-10:30
Calibrating Zero-shot Cross-lingual (Un-)structured Predictions
09:00-09:15
Visual Spatial Description: Controlled Spatial-Oriented Image-to-Text Generation
09:15-09:30
SubeventWriter: Iterative Sub-event Sequence Generation with Coherence Controller
09:30-09:45
Towards a Unified Multi-Dimensional Evaluator for Text Generation
09:45-10:00
Prompt-and-Rerank: A Method for Zero-Shot and Few-Shot Arbitrary Textual Style Transfer with Small Language Models
10:00-10:15
Gradient-based Constrained Sampling from Language Models
10:15-10:30
[TACL] Improving the Domain Adaptation of Retrieval Augmented Generation (RAG) Models for Open-Domain Question-Answering
09:00-09:15
Certified Error Control of Candidate Set Pruning for Two-Stage Relevance Ranking
09:15-09:30
RetroMAE: Pre-Training Retrieval-oriented Language Models Via Masked Auto-Encoder
09:30-09:45
Efficient Document Retrieval by End-to-End Refining and Quantizing BERT Embedding with Contrastive Product Quantization
09:45-10:00
Prompt-Based Meta-Learning For Few-shot Text Classification
10:00-10:15
Generative Multi-hop Retrieval
10:15-10:30
COCO-DR: Combating the Distribution Shift in Zero-Shot Dense Retrieval with Contrastive and Distributionally Robust Learning
09:00-09:15
[INDUSTRY] Grafting Pre-trained Models for Multimodal Headline Generation
09:15-09:30
[INDUSTRY] Named Entity Recognition in Industrial Tables using Tabular Language Models
09:30-09:45
[INDUSTRY] Knowledge Distillation Transfer Sets and their Impact on Downstream NLU Tasks
[INDUSTRY] Iterative Stratified Testing and Measurement for Automated Model Updates
10:00-10:15
[INDUSTRY] Augmenting Operations Research with Auto-Formulation of Optimization Models From Problem Descriptions
10:15-10:30
[INDUSTRY] Distilling Multilingual Transformers into CNNs for Scalable Intent Classification
09:00-09:15
Inducer-tuning: Connecting Prefix-tuning and Adapter-tuning
09:15-09:30
LightEA: A Scalable, Robust, and Interpretable Entity Alignment Framework via Three-view Label Propagation
09:30-09:45
VIRT: Improving Representation-based Text Matching via Virtual Interaction
09:45-10:00
Learning Label Modular Prompts for Text Classification in the Wild
10:00-10:15
COST-EFF: Collaborative Optimization of Spatial and Temporal Efficiency with Slenderized Multi-exit Language Models
10:15-10:30
Training Dynamics for Curriculum Learning: A Study on Monolingual and Cross-lingual NLU
09:00-10:30
Memory-assisted prompt editing to improve GPT-3 after deployment
09:00-10:30
Two is Better than Many? Binary Classification as an Effective Approach to Multi-Choice Question Answering
09:00-10:30
Discovering Differences in the Representation of People using Contextualized Semantic Axes
09:00-10:30
Borrowing Human Senses: Comment-Aware Self-Training for Social Media Multimodal Classification
09:00-10:30
Reflect, Not Reflex: Inference-Based Common Ground Improves Dialogue Response Quality
09:00-10:30
Eliciting Knowledge from Large Pre-Trained Models for Unsupervised Knowledge-Grounded Conversation
09:00-10:30
Discourse Comprehension: A Question Answering Framework to Represent Sentence Connections
09:00-10:30
The Optimal BERT Surgeon: Scalable and Accurate Second-Order Pruning for Large Language Models
09:00-10:30
Syntactic Multi-view Learning for Open Information Extraction
09:00-10:30
DuReader-Retrieval: A Large-scale Chinese Benchmark for Passage Retrieval from Web Search Engine
09:00-10:30
CODER: An efficient framework for improving retrieval through COntextual Document Embedding Reranking
09:00-10:30
Finding Dataset Shortcuts with Grammar Induction
09:00-10:30
SLING: Sino Linguistic Evaluation of Large Language Models
09:00-10:30
Textual Backdoor Attacks Can Be More Harmful via Two Simple Tricks
09:00-10:30
AdaMix: Mixture-of-Adaptations for Parameter-efficient Model Tuning
09:00-10:30
Spectral Probing
09:00-10:30
Cascading Biases: Investigating the Effect of Heuristic Annotation Strategies on Data and Models
09:00-10:30
Quality Scoring of Source Words in Neural Translation Models
09:00-10:30
Language Contamination Helps Explains the Cross-lingual Capabilities of English Pretrained Models
09:00-10:30
PRO-CS : An Instance-Based Prompt Composition Technique for Code-Switched Tasks
09:00-10:30
Multitask Instruction-based Prompting for Fallacy Recognition
09:00-10:30
Unsupervised Non-transferable Text Classification
09:00-10:30
Data-Efficient Playlist Captioning With Musical and Linguistic Knowledge
09:00-10:30
[CL] How Much Does Lookahead Matter for Disambiguation? Partial Arabic Diacritization Case Study
09:00-10:30
[CL] Revise and Resubmit: An Intertextual Model of Text-based Collaboration in Peer Review
09:00-10:30
Re3: Generating Longer Stories With Recursive Reprompting and Revision
09:00-10:30
IndicNLG Benchmark: Multilingual Datasets for Diverse NLG Tasks in Indic Languages
09:00-10:30
You Only Need One Model for Open-domain Question Answering
09:00-10:30
ASQA: Factoid Questions Meet Long-Form Answers
09:00-10:30
ReasTAP: Injecting Table Reasoning Skills During Pre-training via Synthetic Reasoning Examples
09:00-10:30
Knowledge Transfer from Answer Ranking to Answer Generation
09:00-10:30
Pre-training Transformer Models with Sentence-Level Objectives for Answer Sentence Selection
09:00-10:30
Towards Knowledge-Intensive Text-to-SQL Semantic Parsing with Formulaic Knowledge
09:00-10:30
Exploring Document-Level Literary Machine Translation with Parallel Paragraphs from World Literature
09:00-10:30
Making Science Simple: Corpora for the Lay Summarisation of Scientific Literature
09:00-10:30
Generate, Discriminate and Contrast: A Semi-Supervised Sentence Representation Learning Framework
09:00-10:30
Looking at the Overlooked: An Analysis on the Word-Overlap Bias in Natural Language Inference
09:00-10:30
CPL: Counterfactual Prompt Learning for Vision and Language Models
09:00-10:30
MGDoc: Pre-training with Multi-granular Hierarchy for Document Image Understanding
09:00-10:30
Cross-Modal Similarity-Based Curriculum Learning for Image Captioning
09:00-10:30
Textless Speech Emotion Conversion using Discrete & Decomposed Representations
09:00-10:30
X-FACTOR: A Cross-metric Evaluation of Factual Correctness in Abstractive Summarization
09:00-10:30
Unsupervised Opinion Summarisation in the Wasserstein Space
09:00-10:30
Referee: Reference-Free Sentence Summarization with Sharper Controllability through Symbolic Knowledge Distillation
09:00-10:30
Learning to Generate Overlap Summaries through Noisy Synthetic Data
09:00-10:30
Questioning the Validity of Summarization Datasets and Improving Their Factual Consistency
09:00-10:30
The Authenticity Gap in Human Evaluation
09:00-10:30
Towards Robust Numerical Question Answering: Diagnosing Numerical Capabilities of NLP Systems
09:00-10:30
[DEMO] ELEVANT: A Fully Automatic Fine-Grained Entity Linking Evaluation and Analysis Tool
09:00-10:30
[DEMO] Evaluate & Evaluation on the Hub: Better Best Practices for Data and Model Measurements
09:00-10:30
[DEMO] Label Sleuth: From Unlabeled Text to a Classifier in a Few Hours
09:00-10:30
[DEMO] KGxBoard: Explainable and Interactive Leaderboard for Evaluation of Knowledge Graph Completion Models
09:00-10:30
[DEMO] SEAL: Interactive Tool for Systematic Error Analysis and Labeling
11:00-11:15
Transformer Feed-Forward Layers Build Predictions by Promoting Concepts in the Vocabulary Space
11:15-11:30
Interpreting Language Models with Contrastive Explanations
11:30-11:45
Balanced Adversarial Training: Balancing Tradeoffs between Fickleness and Obstinacy in NLP Models
11:45-12:00
DropMix: A Textual Data Augmentation Combining Dropout with Mixup
12:00-12:15
"Will You Find These Shortcuts?" A Protocol for Evaluating the Faithfulness of Input Salience Methods for Text Classification
12:15-12:30
On the Transformation of Latent Space in Fine-Tuned NLP Models
11:00-11:15
Backdoor Attacks in Federated Learning by Rare Embeddings and Gradient Ensembling
11:15-11:30
When Can Transformers Ground and Compose: Insights from Compositional Generalization Benchmarks
11:30-11:45
GammaE: Gamma Embeddings for Logical Queries on Knowledge Graphs
11:45-12:00
Numerical Optimizations for Weighted Low-rank Estimation on Language Models
12:00-12:15
Efficient Nearest Neighbor Search for Cross-Encoder Models using Matrix Factorization
12:15-12:30
Large language models are few-shot clinical information extractors
11:00-11:15
StoryER: Automatic Story Evaluation via Ranking, Rating and Reasoning
Linguistic Corpus Annotation for Automatic Text Simplification Evaluation
11:30-11:45
Near-Negative Distinction: Giving a Second Life to Human Evaluation Datasets
11:45-12:00
Stanceosaurus: Classifying Stance Towards Multicultural Misinformation
12:00-12:15
When FLUE Meets FLANG: Benchmarks and Large Pretrained Language Model for Financial Domain
12:15-12:30
Reproducibility in Computational Linguistics: Is Source Code Enough?
11:00-11:15
Transfer Learning from Semantic Role Labeling to Event Argument Extraction with Template-based Slot Querying
11:15-11:30
Generative Knowledge Graph Construction: A Review
11:30-11:45
Graph-based Model Generation for Few-Shot Relation Extraction
11:45-12:00
A Good Neighbor, A Found Treasure: Mining Treasured Neighbors for Knowledge Graph Entity Typing
12:00-12:15
ReSel: N-ary Relation Extraction from Scientific Text and Tables by Learning to Retrieve and Select
12:15-12:30
MAVEN-ERE: A Unified Large-scale Dataset for Event Coreference, Temporal, Causal, and Subevent Relation Extraction
11:00-11:15
[TACL] Naturalistic Causal Probing for Morpho-Syntax
11:15-11:30
[TACL] Modeling Non-Cooperative Dialogue: Theoretical and Empirical Insights
11:30-11:45
[TACL] Diff-Explainer: Differentiable Convex Optimization for Explainable Multi-hop Inference
11:45-12:00
[TACL] Learning Fair Representations via Rate-Distortion Maximization
12:00-12:15
[TACL] Template-based Abstractive Microblog Opinion Summarisation
12:15-12:30
[TACL] Meta-Learning the Difference: Preparing Large Language Models for Efficient Adaptation
11:00-11:12
Towards Climate Awareness in NLP Research
11:12-11:24
Whose Language Counts as High Quality? Measuring Language Ideologies in Text Data Selection
11:24-11:36
Geographic Citation Gaps in NLP Research
11:36-11:48
[CL] Information Theory-based Compositional Distributional Semantics
11:48-12:00
Extracted BERT Model Leaks More Information than You Think!
12:00-12:12
Exploiting domain-slot related keywords description for Few-Shot Cross-Domain Dialogue State Tracking
12:12-12:24
PRINCE: Prefix-Masked Decoding for Knowledge Enhanced Sequence-to-Sequence Pre-Training
11:00-12:30
Retrieval Augmentation for Commonsense Reasoning: A Unified Approach
11:00-12:30
Rainier: Reinforced Knowledge Introspector for Commonsense Question Answering
11:00-12:30
Empowering the Fact-checkers! Automatic Identification of Claim Spans on Twitter
11:00-12:30
Dealing with Abbreviations in the Slovenian Biographical Lexicon
11:00-12:30
Improving Multi-turn Emotional Support Dialogue Generation with Lookahead Strategy Planning
11:00-12:30
Group is better than individual: Exploiting Label Topologies and Label Relations for Joint Multiple Intent Detection and Slot Filling
11:00-12:30
Information-Theoretic Text Hallucination Reduction for Video-grounded Dialogue
11:00-12:30
Dungeons and Dragons as a Dialog Challenge for Artificial Intelligence
11:00-12:30
An Efficient Memory-Augmented Transformer for Knowledge-Intensive NLP Tasks
11:00-12:30
Leveraging QA Datasets to Improve Generative Data Augmentation
11:00-12:30
An Empirical Study on the Transferability of Transformer Modules in Parameter-efficient Fine-tuning
11:00-12:30
Ethics consideration sections in natural language processing papers
11:00-12:30
An Empirical Study on Finding Spans
11:00-12:30
Simple Questions Generate Named Entity Recognition Datasets
11:00-12:30
Exploring Dual Encoder Architectures for Question Answering
11:00-12:30
Let the CAT out of the bag: Contrastive Attributed explanations for Text
11:00-12:30
Attentional Probe: Estimating a Module’s Functional Potential
11:00-12:30
Predicting Fine-Tuning Performance with Probing
11:00-12:30
BioReader: a Retrieval-Enhanced Text-to-Text Transformer for Biomedical Literature
11:00-12:30
Bernice: A Multilingual Pre-trained Encoder for Twitter
11:00-12:30
ATTEMPT: Parameter-Efficient Multi-task Tuning via Attentional Mixtures of Soft Prompts
11:00-12:30
SocioProbe: What, When, and Where Language Models Learn about Sociodemographics
11:00-12:30
ZeroGen: Efficient Zero-shot Learning via Dataset Generation
11:00-12:30
Entropy- and Distance-Based Predictors From GPT-2 Attention Patterns Predict Reading Times Over and Above GPT-2 Surprisal
11:00-12:30
Discourse Context Predictability Effects in Hindi Word Order
11:00-12:30
Exploration of the Usage of Color Terms by Color-blind Participants in Online Discussion Platforms
11:00-12:30
[TACL] Assessing the capacity of transformer to abstract syntactic representations: a contrastive analysis based on long-distance agreement
11:00-12:30
[INDUSTRY] Zero-Shot Dynamic Quantization for Transformer Inference
11:00-12:30
[INDUSTRY] Prototype-Representations for Training Data Filtering in Weakly-Supervised Information Extraction
11:00-12:30
[INDUSTRY] Entity-level Sentiment Analysis in Contact Center Telephone Conversations
11:00-12:30
[INDUSTRY] QUILL: Query Intent with Large Language Models using Retrieval Augmentation and Multi-stage Distillation
11:00-12:30
PATS: Sensitivity-aware Noisy Learning for Pretrained Language Models
11:00-12:30
Complex Hyperbolic Knowledge Graph Embeddings with Fast Fourier Transform
11:00-12:30
CTL++: Evaluating Generalization on Never-Seen Compositional Patterns of Known Functions, and Compatibility of Neural Representations
11:00-12:30
Adaptive Label Smoothing with Self-Knowledge in Natural Language Generation
11:00-12:30
MM-Align: Learning Optimal Transport-based Alignment Dynamics for Fast and Accurate Inference on Missing Modality Sequences
11:00-12:30
Diverse Parallel Data Synthesis for Cross-Database Adaptation of Text-to-SQL Parsers
11:00-12:30
Normalizing Mutual Information for Robust Adaptive Training for Translation
11:00-12:30
Bilingual Synchronization: Restoring Translational Relationships with Editing Operations
11:00-12:30
Does Joint Training Really Help Cascaded Speech Translation?
11:00-12:30
Discovering Language-neutral Sub-networks in Multilingual Language Models
11:00-12:30
Don’t Stop Fine-Tuning: On Training Regimes for Few-Shot Cross-Lingual Transfer with Multilingual Language Models
11:00-12:30
Improving Low-Resource Languages in Pre-Trained Multilingual Language Models
11:00-12:30
RED-ACE: Robust Error Detection for ASR using Confidence Embeddings
11:00-12:30
Towards Compositional Generalization in Code Search
11:00-12:30
Conditional set generation using Seq2seq models
11:00-12:30
Controlled Text Reduction
11:00-12:30
Break it Down into BTS: Basic, Tiniest Subword Units for Korean
11:00-12:30
Improving Passage Retrieval with Zero-Shot Question Generation
11:00-12:30
Analogical Math Word Problems Solving with Enhanced Problem-Solution Association
11:00-12:30
Super-NaturalInstructions: Generalization via Declarative Instructions on 1600+ NLP Tasks
11:00-12:30
DivEMT: Neural Machine Translation Post-Editing Effort Across Typologically Diverse Languages
11:00-12:30
Human-Machine Collaboration Approaches to Build a Dialogue Dataset for Hate Speech Countering
11:00-12:30
Bloom Library: Multimodal Datasets in 300+ Languages for a Variety of Downstream Tasks
11:00-12:30
DiscoSense: Commonsense Reasoning with Discourse Connectives
11:00-12:30
GraphQ IR: Unifying the Semantic Parsing of Graph Query Languages with One Intermediate Representation
11:00-12:30
QASem Parsing: Text-to-text Modeling of QA-based Semantics
11:00-12:30
Correcting Diverse Factual Errors in Abstractive Summarization via Post-Editing and Language Model Infilling
11:00-12:30
On Parsing as Tagging
11:00-12:30
Structural generalization is hard for sequence-to-sequence models
11:00-12:30
CONDAQA: A Contrastive Reading Comprehension Dataset for Reasoning about Negation
11:00-12:30
Rethinking Style Transformer with Energy-based Interpretation: Adversarial Unsupervised Style Transfer using a Pretrained Model
11:00-12:30
[DEMO] DeepGen: Diverse Search Ad Generation and Real-Time Customization
11:00-12:30
[DEMO] ACCoRD: A Multi-Document Approach to Generating Diverse Descriptions of Scientific Concepts
11:00-12:30
[DEMO] GEMv2: Multilingual NLG Benchmarking in a Single Line of Code
15:30-17:00
Back to the Future: Bidirectional Information Decoupling Network for Multi-turn Dialogue Modeling
15:30-17:00
CGoDial: A Large-Scale Benchmark for Chinese Goal-oriented Dialog Evaluation
15:30-17:00
Supervised Prototypical Contrastive Learning for Emotion Recognition in Conversation
15:30-17:00
Towards Efficient Dialogue Pre-training with Transferable and Interpretable Latent Structure
15:30-17:00
FETA: A Benchmark for Few-Sample Task Transfer in Open-Domain Dialogue
15:30-17:00
IM2: an Interpretable and Multi-category Integrated Metric Framework for Automatic Dialogue Evaluation
15:30-17:00
Neural-based Mixture Probabilistic Query Embedding for Answering FOL queries on Knowledge Graphs
15:30-17:00
Calibrating Student Models for Emotion-related Tasks
15:30-17:00
Automatic Document Selection for Efficient Encoder Pretraining
15:30-17:00
COLD: A Benchmark for Chinese Offensive Language Detection
15:30-17:00
[INDUSTRY] DynaMaR: Dynamic Prompt with Mask Token Representation
15:30-17:00
[INDUSTRY] PENTATRON: PErsonalized coNText-Aware Transformer for Retrieval-based cOnversational uNderstanding
15:30-17:00
Knowledge Prompting in Pre-trained Language Model for Natural Language Understanding
15:30-17:00
Red Teaming Language Models with Language Models
15:30-17:00
Invariant Language Modeling
15:30-17:00
Finding Skill Neurons in Pre-trained Transformer-based Language Models
15:30-17:00
Model Criticism for Long-Form Text Generation
15:30-17:00
[DEMO] EasyNLP: A Comprehensive and Easy-to-use Toolkit for Natural Language Processing
15:30-17:00
[INDUSTRY] Ask-and-Verify: Span Candidate Generation and Verification for Attribute Value Extraction
15:30-17:00
[INDUSTRY] Deploying a Retrieval based Response Model for Task Oriented Dialogues
15:30-17:00
[INDUSTRY] SLATE: A Sequence Labeling Approach for Task Extraction from Free-form Inked Content
15:30-17:00
[INDUSTRY] Meta-learning Pathologies from Radiology Reports using Variance Aware Prototypical Networks
15:30-17:00
[INDUSTRY] Bringing the State-of-the-Art to Customers: A Neural Agent Assistant Framework for Customer Service Support
15:30-17:00
A Unified Positive-Unlabeled Learning Framework for Document-Level Relation Extraction with Different Levels of Labeling
15:30-17:00
Retrieval-Augmented Generative Question Answering for Event Argument Extraction
15:30-17:00
Syntactically Rich Discriminative Training: An Effective Method for Open Information Extraction
15:30-17:00
Improving Event Coreference Resolution Using Document-level and Topic-level Information
15:30-17:00
Modeling Label Correlations for Ultra-Fine Entity Typing with Neural Pairwise Conditional Random Field
15:30-17:00
Query-based Instance Discrimination Network for Relational Triple Extraction
15:30-17:00
Towards Better Document-level Relation Extraction via Iterative Inference
15:30-17:00
Learning Cross-Task Dependencies for Joint Extraction of Entities, Events, Event Arguments, and Relations
15:30-17:00
Entity-centered Cross-document Relation Extraction
15:30-17:00
ConvTrans: Transforming Web Search Sessions for Conversational Dense Retrieval
15:30-17:00
Improving Multi-task Stance Detection with Multi-task Interaction Network
15:30-17:00
Generative Entity Typing with Curriculum Learning
15:30-17:00
Explicit Query Rewriting for Conversational Dense Retrieval
15:30-17:00
Reduce Catastrophic Forgetting of Dense Retrieval Training with Teleportation Negatives
15:30-17:00
An Adaptive Logical Rule Embedding Model for Inductive Reasoning over Temporal Knowledge Graphs
15:30-17:00
[DEMO] DeepKE: A Deep Learning Based Knowledge Extraction Toolkit for Knowledge Base Population
15:30-17:00
Efficient Adversarial Training with Robust Early-Bird Tickets
15:30-17:00
Can Transformers Reason in Fragments of Natural Language?
15:30-17:00
Is the Brain Mechanism for Hierarchical Structure Building Universal Across Languages? An fMRI Study of Chinese and English
15:30-17:00
Conformal Predictor for Improving Zero-Shot Text Classification Efficiency
15:30-17:00
Improving Stability of Fine-Tuning Pretrained Language Models via Component-Wise Gradient Norm Clipping
15:30-17:00
Learning Inter-Entity-Interaction for Few-Shot Knowledge Graph Completion
15:30-17:00
Simplified Graph Learning for Inductive Short Text Classification
15:30-17:00
Interventional Training for Out-Of-Distribution Natural Language Understanding
15:30-17:00
Norm-based Noisy Corpora Filtering and Refurbishing in Neural Machine Translation
15:30-17:00
Helping the Weak Makes You Strong: Simple Multi-Task Learning Improves Non-Autoregressive Translators
15:30-17:00
Modeling Consistency Preference via Lexical Chains for Document-level Neural Machine Translation
15:30-17:00
Breaking the Representation Bottleneck of Chinese Characters: Neural Machine Translation with Stroke Sequence Modeling
15:30-17:00
Increasing Visual Awareness in Multimodal Neural Machine Translation from an Information Theoretic Perspective
15:30-17:00
Adaptive Token-level Cross-lingual Feature Mixing for Multilingual Neural Machine Translation
15:30-17:00
Low-resource Neural Machine Translation with Cross-modal Alignment
15:30-17:00
Multi-level Distillation of Semantic Knowledge for Pre-training Multilingual Language Model
15:30-17:00
Label-aware Multi-level Contrastive Learning for Cross-lingual Spoken Language Understanding
15:30-17:00
[DEMO] BMCook: A Task-agnostic Compression Toolkit for Big Models
15:30-17:00
Distilling Causal Effect from Miscellaneous Other-Class for Continual Named Entity Recognition
15:30-17:00
MedCLIP: Contrastive Learning from Unpaired Medical Images and Text
15:30-17:00
Improving Chinese Spelling Check by Character Pronunciation Prediction: The Effects of Adaptivity and Granularity
15:30-17:00
A Speaker-Aware Co-Attention Framework for Medical Dialogue Information Extraction
15:30-17:00
Affective Knowledge Enhanced Multiple-Graph Fusion Networks for Aspect-based Sentiment Analysis
15:30-17:00
FormLM: Recommending Creation Ideas for Online Forms by Modelling Semantic and Structural Information
15:30-17:00
Mask the Correct Tokens: An Embarrassingly Simple Approach for Error Correction
15:30-17:00
MoSE: Modality Split and Ensemble for Multimodal Knowledge Graph Completion
15:30-17:00
[CL] Effective Approaches to Neural Query Language Identification
15:30-17:00
CapOnImage: Context-driven Dense-Captioning on Image
15:30-17:00
DSM: Question Generation over Knowledge Base via Modeling Diverse Subgraphs with Meta-learner
15:30-17:00
Contrastive Learning enhanced Author-Style Headline Generation
15:30-17:00
Investigating the Robustness of Natural Language Generation from Logical Forms via Counterfactual Samples
15:30-17:00
R2D2: Robust Data-to-Text with Replacement Detection
15:30-17:00
Precisely the Point: Adversarial Augmentations for Faithful and Informative Text Generation
15:30-17:00
Towards Inter-character Relationship-driven Story Generation
15:30-17:00
ProofInfer: Generating Proof via Iterative Hierarchical Inference
15:30-17:00
[DEMO] SUMMARY WORKBENCH: Unifying Application and Evaluation of Text Summarization Models
15:30-17:00
[DEMO] BotSIM: An End-to-End Bot Simulation Framework for Commercial Task-Oriented Dialog Systems
15:30-17:00
[DEMO] TextBox 2.0: A Text Generation Library with Pre-trained Language Models
15:30-17:00
Improving Tokenisation by Alternative Treatment of Spaces
15:30-17:00
Capturing Global Structural Information in Long Document Question Answering with Compressive Graph Selector Network
15:30-17:00
DRLK: Dynamic Hierarchical Reasoning with Language Model and Knowledge Graph for Question Answering
15:30-17:00
TIARA: Multi-grained Retrieval for Robust Question Answering over Large Knowledge Base
15:30-17:00
Rethinking Multi-Modal Alignment in Multi-Choice VideoQA from Feature and Sample Perspectives
15:30-17:00
A Second Wave of UD Hebrew Treebanking and Cross-Domain Parsing
15:30-17:00
MetaLogic: Logical Reasoning Explanations with Fine-Grained Structure
15:30-17:00
Rethinking the Authorship Verification Experimental Setups
15:30-17:00
Hierarchical Multi-Label Classification of Scientific Documents
15:30-17:00
CISLR: Corpus for Indian Sign Language Recognition
15:30-17:00
[CL] The Text Anonymization Benchmark (TAB): A Dedicated Corpus and Evaluation Framework for Text Anonymization
15:30-17:00
Sentence Representation Learning with Generative Objective rather than Contrastive Objective
15:30-17:00
Curriculum Learning Meets Weakly Supervised Multimodal Correlation Learning
15:30-17:00
Efficient Nearest Neighbor Emotion Classification with BERT-whitening
15:30-17:00
AEG: Argumentative Essay Generation via A Dual-Decoder Model with Content Planning
15:30-17:00
Symptom Identification for Interpretable Detection of Multiple Mental Disorders on Social Media
15:30-17:00
[DEMO] AnEMIC: A Framework for Benchmarking ICD Coding Models
15:30-17:00
[DEMO] MedConQA: Medical Conversational Question Answering System based on Knowledge Graphs
15:30-17:00
LVP-M3: Language-aware Visual Prompt for Multilingual Multimodal Machine Translation
15:30-17:00
UniGeo: Unifying Geometry Logical Reasoning via Reformulating Mathematical Expression
15:30-17:00
An Anchor-based Relative Position Embedding Method for Cross-Modal Tasks
15:30-17:00
A Span-based Multimodal Variational Autoencoder for Semi-supervised Multimodal Named Entity Recognition
15:30-17:00
Towards Unifying Reference Expression Generation and Comprehension
15:30-17:00
Speaker Overlap-aware Neural Diarization for Multi-party Meeting Analysis
15:30-17:00
Extending Phrase Grounding with Pronouns in Visual Dialogues
15:30-17:00
Distilled Dual-Encoder Model for Vision-Language Understanding
15:30-17:00
PEVL: Position-enhanced Pre-training and Prompt Tuning for Vision-language Models
15:30-17:00
RACE: Retrieval-augmented Commit Message Generation
15:30-17:00
Leveraging Locality in Abstractive Text Summarization
15:30-17:00
Assist Non-native Viewers: Multimodal Cross-Lingual Summarization for How2 Videos
15:30-17:00
ClidSum: A Benchmark Dataset for Cross-Lingual Dialogue Summarization
15:30-17:00
[TACL] A Survey on Cross-Lingual Summarization
15:30-17:00
PASTA: Table-Operations Aware Fact Verification via Sentence-Table Cloze Pre-training
15:30-17:00
Do Children Texts Hold The Key To Commonsense Knowledge?
15:30-17:00
Understanding ME? Multimodal Evaluation for Fine-grained Visual Commonsense
15:30-17:00
Multi-Label Intent Detection via Contrastive Task Specialization of Sentence Encoders
15:30-17:00
[TACL] Evaluating Attribution in Dialogue Systems: The BEGIN Benchmark
15:30-17:00
Is a Question Decomposition Unit All We Need?
15:30-17:00
EdgeFormer: A Parameter-Efficient Transformer for On-Device Seq2seq Generation
15:30-17:00
MuRAG: Multimodal Retrieval-Augmented Generator for Open Question Answering over Images and Text
15:30-17:00
Debiasing Pretrained Text Encoders by Paying Attention to Paying Attention
15:30-17:00
SetGNER: General Named Entity Recognition as Entity Set Generation
15:30-17:00
Does Your Model Classify Entities Reasonably? Diagnosing and Mitigating Spurious Correlations in Entity Typing
15:30-17:00
Decoding a Neural Retriever’s Latent Space for Query Suggestion
15:30-17:00
GPS: Genetic Prompt Search for Efficient Few-Shot Learning
15:30-17:00
Continual Training of Language Models for Few-Shot Learning
15:30-17:00
WeTS: A Benchmark for Translation Suggestion
15:30-17:00
T-Modules: Translation Modules for Zero-Shot Cross-Modal Machine Translation
15:30-17:00
Joint Completion and Alignment of Multilingual Knowledge Graphs
15:30-17:00
BERT in Plutarch’s Shadows
15:30-17:00
Composing Ci with Reinforced Non-autoregressive Text Generation
15:30-17:00
ConvFinQA: Exploring the Chain of Numerical Reasoning in Conversational Finance Question Answering
15:30-17:00
Uni-Parser: Unified Semantic Parser for Question Answering on Knowledge Base and Database
15:30-17:00
GENIE: Toward Reproducible and Standardized Human Evaluation for Text Generation
15:30-17:00
Open World Classification with Adaptive Negative Samples
15:30-17:00
FLUTE: Figurative Language Understanding through Textual Explanations
15:30-17:00
Breakpoint Transformers for Modeling and Tracking Intermediate Beliefs
15:30-17:00
Sentiment-Aware Word and Sentence Level Pre-training for Sentiment Analysis
15:30-17:00
Improving Aspect Sentiment Quad Prediction via Template-Order Data Augmentation
15:30-17:00
CTRLsum: Towards Generic Controllable Text Summarization
Sunday 11th December
09:00-09:15
Using Commonsense Knowledge to Answer Why-Questions
09:15-09:30
Maieutic Prompting: Logically Consistent Reasoning with Recursive Explanations
09:30-09:45
Language Models of Code are Few-Shot Commonsense Learners
09:45-10:00
Enhancing Self-Consistency and Performance of Pre-Trained Language Models through Natural Language Inference
10:00-10:15
EvEntS ReaLM: Event Reasoning of Entity States via Language Models
10:15-10:30
GeoMLAMA: Geo-Diverse Commonsense Probing on Multilingual Pre-Trained Language Models
09:00-09:15
The Importance of Being Parameters: An Intra-Distillation Method for Serious Gains
09:15-09:30
Non-Parametric Domain Adaptation for End-to-End Speech Translation
09:30-09:45
Information-Transport-based Policy for Simultaneous Translation
09:45-10:00
Multilingual Machine Translation with Hyper-Adapters
10:00-10:15
Continual Learning of Neural Machine Translation within Low Forgetting Risk Regions
10:15-10:30
Distill The Image to Nowhere: Inversion Knowledge Distillation for Multimodal Machine Translation
09:00-09:12
A Multilingual Perspective Towards the Evaluation of Attribution Methods in Natural Language Inference
09:12-09:24
Robustness of Demonstration-based Learning Under Limited Data Scenario
09:24-09:36
Entailer: Answering Questions with Faithful and Truthful Chains of Reasoning
09:36-09:48
Revisiting Parameter-Efficient Tuning: Are We Really There Yet?
09:48-10:00
Word Order Matters When You Increase Masking
10:00-10:12
Stop Measuring Calibration When Humans Disagree
10:12-10:24
Are Hard Examples also Harder to Explain? A Study with Human and Model-Generated Explanations
09:00-09:15
Bilingual Lexicon Induction for Low-Resource Languages using Graph Matching via Optimal Transport
09:15-09:30
Zero-Shot Text Classification with Self-Training
09:30-09:45
Fine-grained Category Discovery under Coarse-grained supervision with Hierarchical Weighted Self-contrastive Learning
09:45-10:00
Learning Instructions with Unlabeled Data for Zero-Shot Cross-Task Generalization
10:00-10:15
Learning to Adapt to Low-Resource Paraphrase Generation
10:15-10:30
ToKen: Task Decomposition and Knowledge Infusion for Few-Shot Hate Speech Detection
09:00-09:15
[INDUSTRY] Improving Precancerous Case Characterization via Transformer-based Ensemble Learning
09:15-09:30
[INDUSTRY] Unsupervised Term Extraction for Highly Technical Domains
09:30-09:45
[INDUSTRY] CoCoID: Learning Contrastive Representations and Compact Clusters for Semi-Supervised Intent Discovery
09:45-10:00
[INDUSTRY] Automatic Scene-based Topic Channel Construction System for E-Commerce
10:00-10:15
[INDUSTRY] Gaining Insights into Unrecognized User Utterances in Task-Oriented Dialog Systems
10:15-10:30
[INDUSTRY] Towards Need-Based Spoken Language Understanding Model Updates: What Have We Learned?
09:00-09:15
Metric-guided Distillation: Distilling Knowledge from the Metric to Ranker and Retriever for Generative Commonsense Reasoning
09:15-09:30
Segmenting Numerical Substitution Ciphers
09:30-09:45
Deconfounding Legal Judgment Prediction for European Court of Human Rights Cases Towards Better Alignment with Experts
09:45-10:00
PLM-based World Models for Text-based Games
10:00-10:15
ConReader: Exploring Implicit Relations in Contracts for Contract Clause Extraction
10:15-10:30
[TACL] Compositional Generalization in Multilingual Semantic Parsing over Wikidata
09:00-10:30
Unifying Data Perspectivism and Personalization: An Application to Social Norms
09:00-10:30
Offer a Different Perspective: Modeling the Belief Alignment of Arguments in Multi-party Debates
09:00-10:30
Less is More: Summary of Long Instructions is Better for Program Synthesis
09:00-10:30
HashFormers: Towards Vocabulary-independent Pre-trained Transformers
09:00-10:30
AMAL: Meta Knowledge-Driven Few-Shot Adapter Learning
09:00-10:30
LittleBird: Efficient Faster & Longer Transformer for Question Answering
09:00-10:30
Understanding and Improving Knowledge Distillation for Quantization Aware Training of Large Transformer Encoders
09:00-10:30
Tutoring Helps Students Learn Better: Improving Knowledge Distillation for BERT with Tutor Network
09:00-10:30
NewsClaims: A New Benchmark for Claim Detection from News with Attribute Knowledge
09:00-10:30
POQue: Asking Participant-specific Outcome Questions for a Deeper Understanding of Complex Events
09:00-10:30
Revisiting DocRED - Addressing the False Negative Problem in Relation Extraction
09:00-10:30
AfriCLIRMatrix: Enabling Cross-Lingual Information Retrieval for African Languages
09:00-10:30
A Dataset for Hyper-Relational Extraction and a Cube-Filling Approach
09:00-10:30
A Fine-grained Chinese Software Privacy Policy Dataset for Sequence Labeling and Regulation Compliant Identification
09:00-10:30
Mitigating Data Sparsity for Short Text Topic Modeling by Topic-Semantic Contrastive Learning
09:00-10:30
WeDef: Weakly Supervised Backdoor Defense for Text Classification
09:00-10:30
Pseudo-Relevance for Enhancing Document Representation
09:00-10:30
Logical Reasoning with Span-Level Predictions for Interpretable and Robust NLI Models
09:00-10:30
Faithful Knowledge Graph Explanations in Commonsense Question Answering
09:00-10:30
Improving Embeddings Representations for Comparing Higher Education Curricula: A Use Case in Computing
09:00-10:30
SPE: Symmetrical Prompt Enhancement for Fact Probing
09:00-10:30
Revisiting Pre-trained Language Models and their Evaluation for Arabic Natural Language Processing
09:00-10:30
On Measuring the Intrinsic Few-Shot Hardness of Datasets
09:00-10:30
Mixture of Attention Heads: Selecting Attention Heads Per Token
09:00-10:30
Transformer-based Entity Typing in Knowledge Graphs
09:00-10:30
Debiasing Masks: A New Framework for Shortcut Mitigation in NLU
09:00-10:30
MT-GenEval: A Counterfactual and Contextual Dataset for Evaluating Gender Accuracy in Machine Translation
09:00-10:30
Multimodal Robustness for Neural Machine Translation
09:00-10:30
Disentangling Uncertainty in Machine Translation Evaluation
09:00-10:30
PreQuEL: Quality Estimation of Machine Translation Outputs in Advance
10:30
[INDUSTRY] Dense Feature Memory Augmented Transformers for COVID-19 Vaccination Search Classification
09:00-10:30
[INDUSTRY] Deploying Unified BERT Moderation Model for E-Commerce Reviews
09:00-10:30
[INDUSTRY] End-to-End Speech to Intent Prediction to improve E-commerce Customer Support Voicebot in Hindi and English
09:00-10:30
Analyzing the Mono- and Cross-Lingual Pretraining Dynamics of Multilingual Language Models
09:00-10:30
On the Calibration of Massively Multilingual Language Models
09:00-10:30
Synergy with Translation Artifacts for Training and Inference in Multilingual Tasks
09:00-10:30
Hyper-X: A Unified Hypernetwork for Multi-Task Multilingual Transfer
09:00-10:30
SLICER: Sliced Fine-Tuning for Low-Resource Cross-Lingual Transfer for Named Entity Recognition
09:00-10:30
Natural Logic-guided Autoregressive Multi-hop Document Retrieval for Fact Verification
09:00-10:30
Topical Segmentation of Spoken Narratives: A Test Case on Holocaust Survivor Testimonies
09:00-10:30
Generalizing over Long Tail Concepts for Medical Term Normalization
09:00-10:30
Neighborhood Contrastive Learning for Scientific Document Representations with Citation Embeddings
09:00-10:30
Discourse-Aware Soft Prompting for Text Generation
09:00-10:30
Not to Overfit or Underfit the Source Domains? An Empirical Study of Domain Generalization in Question Answering
09:00-10:30
Momentum Contrastive Pre-training for Question Answering
09:00-10:30
CEFR-Based Sentence Difficulty Annotation and Assessment
09:00-10:30
IDK-MRC: Unanswerable Questions for Indonesian Machine Reading Comprehension
09:00-10:30
Open-domain Video Commentary Generation
09:00-10:30
EUR-Lex-Sum: A Multi- and Cross-lingual Dataset for Long-form Summarization in the Legal Domain
09:00-10:30
The Aligned Multimodal Movie Treebank: An audio, video, dependency-parse treebank
09:00-10:30
KOLD: Korean Offensive Language Dataset
09:00-10:30
ECTSum: A New Benchmark Dataset For Bullet Point Summarization of Long Earnings Call Transcripts
09:00-10:30
Inductive Relation Prediction with Logical Reasoning Using Contrastive Representations
09:00-10:30
IndicXNLI: Evaluating Multilingual Inference for Indian Languages
09:00-10:30
Generating Literal and Implied Subquestions to Fact-check Complex Claims
09:00-10:30
Entity-Focused Dense Passage Retrieval for Outside-Knowledge Visual Question Answering
09:00-10:30
Don’t Copy the Teacher: Data and Model Challenges in Embodied Dialogue
09:00-10:30
Scientific Paper Extractive Summarization Enhanced by Citation Graphs
09:00-10:30
Analyzing and Evaluating Faithfulness in Dialogue Summarization
09:00-10:30
How "Multi" is Multi-Document Summarization?
09:00-10:30
Factorizing Content and Budget Decisions in Abstractive Summarization of Long Documents
09:00-10:30
Evaluating and Improving Factuality in Multimodal Abstractive Summarization
09:00-10:30
IsoVec: Controlling the Relative Isomorphism of Word Embedding Spaces
09:00-10:30
[DEMO] SPEAR : Semi-supervised Data Programming in Python
09:00-10:30
[DEMO] FALTE: A Toolkit for Fine-grained Annotation for Long Text Evaluation
09:00-10:30
[DEMO] ALToolbox: A Set of Tools for Active Learning Annotation of Natural Language Texts
09:00-10:30
[DEMO] A Pipeline for Generating, Annotating and Employing Synthetic Data for Real World Question Answering
11:00-11:15
SpeechUT: Bridging Speech and Text with Hidden-Unit for Encoder-Decoder Based Speech-Text Pre-training
11:15-11:30
Can Visual Context Improve Automatic Speech Recognition for an Embodied Agent?
11:30-11:45
Why is Winoground Hard? Investigating Failures in Visuolinguistic Compositionality
11:45-12:00
Do Vision-and-Language Transformers Learn Grounded Predicate-Noun Dependencies?
12:00-12:15
[TACL] Learning English with Peppa Pig
12:15-12:30
[TACL] Draw Me a Flower: Processing and Grounding Abstraction in Natural Language
11:00-11:15
Learning to Decompose: Hypothetical Question Decomposition Based on Comparable Texts
11:15-11:30
TaCube: Pre-computing Data Cubes for Answering Numerical-Reasoning Questions over Tabular Data
11:30-11:45
Rich Knowledge Sources Bring Complex Knowledge Conflicts: Recalibrating Models to Reflect Conflicting Evidence
11:45-12:00
QA Domain Adaptation using Hidden Space Augmentation and Self-Supervised Contrastive Adaptation
12:00-12:15
Retrieval as Attention: End-to-end Learning of Retrieval and Reading within a Single Transformer
12:15-12:30
Generating Information-Seeking Conversations from Unlabeled Documents
11:00-11:15
MetaASSIST: Robust Dialogue State Tracking with Meta Learning
11:15-11:30
Watch the Neighbors: A Unified K-Nearest Neighbor Contrastive Learning Framework for OOD Intent Discovery
11:30-11:45
Counterfactual Data Augmentation via Perspective Transition for Open-Domain Dialogues
11:45-12:00
There Is No Standard Answer: Knowledge-Grounded Dialogue Generation with Adversarial Activated Multi-Reference Learning
12:00-12:15
D4: a Chinese Dialogue Dataset for Depression-Diagnosis-Oriented Chat
12:15-12:30
Navigating Connected Memories with a Task-oriented Dialog System
11:00-11:15
Entity Extraction in Low Resource Domains with Selective Pre-training of Large Language Models
11:15-11:30
Multilingual Relation Classification via Efficient and Effective Prompting
11:30-11:45
Fine-grained Contrastive Learning for Relation Extraction
11:45-12:00
SQUIRE: A Sequence-to-sequence Framework for Multi-hop Knowledge Graph Reasoning
12:00-12:15
Style Transfer as Data Augmentation: A Case Study on Named Entity Recognition
12:15-12:30
Rescue Implicit and Long-tail Cases: Nearest Neighbor Relation Extraction
11:00-11:15
[TACL] Causal Inference in Natural Language Processing: Estimation, Prediction, Interpretation and Beyond
11:15-11:30
[TACL] Multi-task Active Learning for Pre-trained Transformer-based Models
11:30-11:45
[TACL] Saturated Transformers are Constant-Depth Threshold Circuits
11:45-12:00
[TACL] Unit Tests for Concepts in Neural Networks
12:00-12:15
[TACL] On the Role of Negative Precedent in Legal Outcome Prediction
12:15-12:30
[TACL] Typical Decoding for Natural Language Generation
11:00-11:15
Unsupervised Boundary-Aware Language Model Pretraining for Chinese Sequence Labeling
11:15-11:30
SynGEC: Syntax-Enhanced Grammatical Error Correction with a Tailored GEC-Oriented Parser
11:30-11:45
Unbiased and Efficient Sampling of Dependency Trees
11:45-12:00
A Comprehensive Comparison of Neural Networks as Cognitive Models of Inflection
12:00-12:15
[TACL] Morphology Without Borders: Clause-Level Morphology
12:15-12:30
[TACL] Transformer Grammars: Augmenting Transformer Language Models with Syntactic Inductive Biases at Scale
11:00-12:30
CDialog: A Multi-turn Covid-19 Conversation Dataset for Entity-Aware Dialog Generation
11:00-12:30
Efficient Pre-training of Masked Language Model via Concept-based Curriculum Masking
11:00-12:30
Meta-Learning Fast Weight Language Models
12:30
ArtELingo: A Million Emotion Annotations of WikiArt with Emphasis on Diversity over Language and Culture
11:00-12:30
Late Fusion with Triplet Margin Objective for Multimodal Ideology Prediction and Analysis
11:00-12:30
Differentiable Data Augmentation for Contrastive Sentence Representation Learning
11:00-12:30
Balancing out Bias: Achieving Fairness Through Balanced Training
11:00-12:30
Coordinated Topic Modeling
11:00-12:30
Large Dual Encoders Are Generalizable Retrievers
11:00-12:30
ADDMU: Detection of Far-Boundary Adversarial Examples with Data and Model Uncertainty Estimation
11:00-12:30
Does Self-Rationalization Improve Robustness to Spurious Correlations?
11:00-12:30
Better Hit the Nail on the Head than Beat around the Bush: Removing Protected Attributes with a Single Projection
11:00-12:30
Training Language Models with Memory Augmentation
11:00-12:30
TemporalWiki: A Lifelong Benchmark for Training and Evaluating Ever-Evolving Language Models
11:00-12:30
Prompting ELECTRA: Few-Shot Learning with Discriminative Pre-Trained Models
11:00-12:30
Context Limitations Make Neural Language Models More Human-Like
11:00-12:30
The better your Syntax, the better your Semantics? Probing Pretrained Language Models for the English Comparative Correlative
11:00-12:30
Continued Pretraining for Better Zero- and Few-Shot Promptability
11:00-12:30
Hierarchical Phrase-Based Sequence-to-Sequence Learning
11:00-12:30
Variational Autoencoder with Disentanglement Priors for Low-Resource Task-Specific Natural Language Generation
11:00-12:30
Non-Autoregressive Neural Machine Translation: A Call for Clarity
11:00-12:30
When does Parameter-Efficient Transfer Learning Work for Machine Translation?
11:00-12:30
SMaLL-100: Introducing Shallow Multilingual Machine Translation Model for Low-Resource Languages
11:00-12:30
Specializing Multi-domain NMT via Penalizing Low Mutual Information
11:00-12:30
Federated Meta-Learning for Emotion and Sentiment Aware Multi-modal Complaint Identification
11:00-12:30
PLOG: Table-to-Logic Pretraining for Logical Table-to-Text Generation
11:00-12:30
Help me write a Poem - Instruction Tuning as a Vehicle for Collaborative Poetry Writing
11:00-12:30
Video Question Answering: Datasets, Algorithms and Challenges
11:00-12:30
Exploring the Secrets Behind the Learning Difficulty of Meaning Representations for Semantic Parsing
11:00-12:30
Are representations built from the ground up? An empirical examination of local composition in language models
11:00-12:30
PCL: Peer-Contrastive Learning with Diverse Augmentations for Unsupervised Sentence Embeddings
11:00-12:30
ULN: Towards Underspecified Vision-and-Language Navigation
11:00-12:30
Summarizing Community-based Question-Answer Pairs
11:00-12:30
Abstractive Summarization Guided by Latent Hierarchical Document Structure
11:00-12:30
SentBS: Sentence-level Beam Search for Controllable Summarization
11:00-12:30
Missing Counter-Evidence Renders NLP Fact-Checking Unrealistic for Misinformation
11:00-12:30
Towards Pragmatic Production Strategies for Natural Language Generation Tasks
11:00-12:30
Neural Theory-of-Mind? On the Limits of Social Intelligence in Large LMs
11:00-12:30
Perturbation Augmentation for Fairer NLP
11:00-12:30
MABEL: Attenuating Gender Bias using Textual Entailment Data
11:00-12:30
Cross-stitching Text and Knowledge Graph Encoders for Distantly Supervised Relation Extraction
11:00-12:30
"Covid vaccine is against Covid but Oxford vaccine is made at Oxford!" Semantic Interpretation of Proper Noun Compounds
11:00-12:30
[INDUSTRY] A Comprehensive Evaluation of Biomedical Entity-centric Search
11:00-12:30
[INDUSTRY] Domain Adaptation of Machine Translation with Crowdworkers
11:00-12:30
[INDUSTRY] Biomedical NER for the Enterprise with Distillated BERN2 and the Kazu Framework
11:00-12:30
[INDUSTRY] Topic Modeling by Clustering Language Model Embeddings: Human Validation on an Industry Dataset
11:00-12:30
Beyond prompting: Making Pre-trained Language Models Better Zero-shot Learners by Clustering Representations
11:00-12:30
[DEMO] KeywordScape: Visual Document Exploration using Contextualized Keyword Embeddings
11:00-12:30
[DEMO] Arabic Word-level Readability Visualization for Assisted Text Simplification
11:00-12:30
[DEMO] LogiTorch: A PyTorch-based library for logical reasoning on natural language
11:00-12:30
[DEMO] Paraphrastic Representations at Scale
11:00-12:30
[DEMO] KGI: An Integrated Framework for Knowledge Intensive Language Tasks
15:30-17:00
Graph Hawkes Transformer for Extrapolated Reasoning on Temporal Knowledge Graphs
15:30-17:00
ACENet: Attention Guided Commonsense Reasoning on Hybrid Knowledge Graph
15:30-17:00
Q-TOD: A Query-driven Task-oriented Dialogue System
15:30-17:00
Dial2vec: Self-Guided Contrastive Learning of Unsupervised Dialogue Embeddings
15:30-17:00
Enhancing Joint Multiple Intent Detection and Slot Filling with Global Intent-Slot Co-occurrence
15:30-17:00
IRRGN: An Implicit Relational Reasoning Graph Network for Multi-turn Response Selection
15:30-17:00
Concadia: Towards Image-Based Text Generation with a Purpose
15:30-17:00
SpanProto: A Two-stage Span-based Prototypical Network for Few-shot Named Entity Recognition
15:30-17:00
RelU-Net: Syntax-aware Graph U-Net for Relational Triple Extraction
15:30-17:00
Wider & Closer: Mixture of Short-channel Distillers for Zero-shot Cross-lingual Named Entity Recognition
15:30-17:00
Learning Robust Representations for Continual Relation Extraction via Adversarial Class Augmentation
15:30-17:00
UniRel: Unified Representation and Interaction for Joint Relational Triple Extraction
15:30-17:00
MetaTKG: Learning Evolutionary Meta-Knowledge for Temporal Knowledge Graph Reasoning
15:30-17:00
IELM: An Open Information Extraction Benchmark for Pre-Trained Language Models
15:30-17:00
Predicting Prerequisite Relations for Unseen Concepts
15:30-17:00
Boosting Document-Level Relation Extraction by Mining and Injecting Logical Rules
15:30-17:00
Towards relation extraction from speech
15:30-17:00
CodeRetriever: A Large Scale Contrastive Pre-Training Method for Code Search
15:30-17:00
Exploring Representation-level Augmentation for Code Search
15:30-17:00
Cross-Linguistic Syntactic Difference in Multilingual BERT: How Good is It and How Does It Affect Transfer?
15:30-17:00
Learning to Explain Selectively: A Case Study on Question Answering
15:30-17:00
ROSE: Robust Selective Fine-tuning for Pre-trained Language Models
15:30-17:00
COPEN: Probing Conceptual Knowledge in Pre-trained Language Models
15:30-17:00
Zero-Shot Learners for Natural Language Understanding via a Unified Multiple Choice Perspective
15:30-17:00
Character-level White-Box Adversarial Attacks against Transformers via Attachable Subwords Substitution
15:30-17:00
XPrompt: Exploring the Extreme of Prompt Tuning
15:30-17:00
Instance Regularization for Discriminative Language Model Pre-training
15:30-17:00
A Survey of Active Learning for Natural Language Processing
15:30-17:00
The Devil in Linear Transformer
15:30-17:00
STGN: an Implicit Regularization Method for Learning with Noisy Labels in Natural Language Processing
15:30-17:00
Candidate Soups: Fusing Candidate Results Improves Translation Quality for Non-Autoregressive Translation
15:30-17:00
Towards Robust k-Nearest-Neighbor Machine Translation
15:30-17:00
Unifying the Convergences in Multilingual Neural Machine Translation
15:30-17:00
Hypoformer: Hybrid Decomposition Transformer for Edge-friendly Neural Machine Translation
15:30-17:00
Reorder and then Parse, Fast and Accurate Discontinuous Constituency Parsing
15:30-17:00
[INDUSTRY] A Stacking-based Efficient Method for Toxic Language Detection on Live Streaming Chat
15:30-17:00
[INDUSTRY] Consultation Checklists: Standardising the Human Evaluation of Medical Note Generation
15:30-17:00
[INDUSTRY] Controlled Language Generation for Language Learning Items
15:30-17:00
[INDUSTRY] Fact Checking Machine Generated Text with Dependency Trees
15:30-17:00
[INDUSTRY] Distinguish Sense from Nonsense: Out-of-Scope Detection for Virtual Assistants
15:30-17:00
[INDUSTRY] PLATO-Ad: A Unified Advertisement Text Generation Framework with Multi-Task Prompt Learning
15:30-17:00
[DEMO] stopes - Modular Machine Translation Pipelines
15:30-17:00
Empowering Dual-Encoder with Query Generator for Cross-Lingual Dense Retrieval
15:30-17:00
Toward the Limitation of Code-Switching in Cross-Lingual Transfer
15:30-17:00
Enhancing Multilingual Language Model with Massive Multilingual Knowledge Triples
15:30-17:00
Zero-shot Cross-lingual Transfer of Prompt-based Tuning with a Unified Multilingual Prompt
15:30-17:00
Open-Topic False Information Detection on Social Networks with Contrastive Adversarial Learning
15:30-17:00
A Joint Learning Framework for Restaurant Survival Prediction and Explanation
15:30-17:00
Towards Multi-Modal Sarcasm Detection via Hierarchical Congruity Modeling with Knowledge Enhancement
15:30-17:00
MetaFill: Text Infilling for Meta-Path Generation on Heterogeneous Information Networks
15:30-17:00
TeleMelody: Lyric-to-Melody Generation with a Template-Based Two-Stage Method
15:30-17:00
Federated Model Decomposition with Private Vocabulary for Text Classification
15:30-17:00
"It’s Not Just Hate": A Multi-Dimensional Perspective on Detecting Harmful Speech Online
15:30-17:00
Semantic Novelty Detection and Characterization in Factual Text Involving Named Entities
15:30-17:00
SHARE: a System for Hierarchical Assistive Recipe Editing
15:30-17:00
A Federated Approach to Predicting Emojis in Hindi Tweets
15:30-17:00
PAR: Political Actor Representation Learning with Social Context and Expert Knowledge
15:30-17:00
Tiny-Attention Adapter: Contexts Are More Important Than the Number of Parameters
15:30-17:00
AdapterShare: Task Correlation Modeling with Adapter Differentiation
15:30-17:00
Rethinking Task-Specific Knowledge Distillation: Contextualized Corpus as Better Textbook
15:30-17:00
Counterfactual Recipe Generation: Exploring Compositional Generalization in a Realistic Scenario
15:30-17:00
DisCup: Discriminator Cooperative Unlikelihood Prompt-tuning for Controllable Text Generation
15:30-17:00
Evaluating Parameter Efficient Learning for Generation
15:30-17:00
Revisiting Grammatical Error Correction Evaluation and Beyond
15:30-17:00
Long Text Generation with Topic-aware Discrete Latent Variable Model
15:30-17:00
[DEMO] LM-Debugger: An Interactive Tool for Inspection and Intervention in Transformer-Based Language Models
15:30-17:00
Evade the Trap of Mediocrity: Promoting Diversity and Novelty in Text Generation via Concentrating Attention
15:30-17:00
KECP: Knowledge Enhanced Contrastive Prompting for Few-shot Extractive Question Answering
15:30-17:00
RLET: A Reinforcement Learning Based Approach for Explainable QA with Entailment Trees
15:30-17:00
UniRPG: Unified Discrete Reasoning over Table and Text as Program Generation
15:30-17:00
MUSIED: A Benchmark for Event Detection from Multi-Source Heterogeneous Informal Texts
15:30-17:00
Reproducibility Issues for BERT-based Evaluation Metrics
15:30-17:00
Context Matters for Image Descriptions for Accessibility: Challenges for Referenceless Evaluation Metrics
15:30-17:00
Transfer Learning with Synthetic Corpora for Spatial Role Labeling and Reasoning
15:30-17:00
arXivEdits: Understanding the Human Revision Process in Scientific Writing
15:30-17:00
JDDC 2.1: A Multimodal Chinese Dialogue Dataset with Joint Tasks of Query Rewriting, Response Generation, Discourse Parsing, and Summarization
15:30-17:00
MEE: A Novel Multilingual Event Extraction Dataset
15:30-17:00
Just Fine-tune Twice: Selective Differential Privacy for Large Language Models
15:30-17:00
R2F: A General Retrieval, Reading and Fusion Framework for Document-level Natural Language Inference
15:30-17:00
RASAT: Integrating Relational Structures into Pretrained Seq2Seq Model for Text-to-SQL
15:30-17:00
Understanding Jargon: Combining Extraction and Generation for Definition Modeling
15:30-17:00
Exploiting Global and Local Hierarchies for Hierarchical Text Classification
15:30-17:00
Learning Semantic Textual Similarity via Topic-informed Discrete Latent Variables
15:30-17:00
Retrofitting Multilingual Sentence Embeddings with Abstract Meaning Representation
15:30-17:00
DEER: Descriptive Knowledge Graph for Explaining Entity Relationships
15:30-17:00
COM-MRC: A COntext-Masked Machine Reading Comprehension Framework for Aspect Sentiment Triplet Extraction
15:30-17:00
CEM: Machine-Human Chatting Handoff via Causal-Enhance Module
15:30-17:00
Face-Sensitive Image-to-Emotional-Text Cross-modal Translation for Multimodal Aspect-based Sentiment Analysis
15:30-17:00
A Span-level Bidirectional Network for Aspect Sentiment Triplet Extraction
15:30-17:00
Generative Data Augmentation with Contrastive Learning for Zero-Shot Stance Detection
15:30-17:00
Text Style Transferring via Adversarial Masking and Styled Filling
15:30-17:00
Generative Entity-to-Entity Stance Detection with Knowledge Graph Augmentation
15:30-17:00
A Simple Contrastive Learning Framework for Interactive Argument Pair Identification via Argument-Context Extraction
15:30-17:00
[DEMO] CogKTR: A Knowledge-Enhanced Text Representation Toolkit for Natural Language Understanding
15:30-17:00
[DEMO] SynKB: Semantic Search for Synthetic Procedures
15:30-17:00
RelCLIP: Adapting Language-Image Pretraining for Visual Relationship Detection via Relational Contrastive Learning
15:30-17:00
Discrete Cross-Modal Alignment Enables Zero-Shot Speech Translation
15:30-17:00
GHAN: Graph-Based Hierarchical Aggregation Network for Text-Video Retrieval
15:30-17:00
Open-Domain Sign Language Translation Learned from Online Video
15:30-17:00
SEMGraph: Incorporating Sentiment Knowledge and Eye Movement into Graph Model for Sentiment Analysis
15:30-17:00
Contrastive Learning with Expectation-Maximization for Weakly Supervised Phrase Grounding
15:30-17:00
Weakly-Supervised Temporal Article Grounding
15:30-17:00
FaD-VLP: Fashion Vision-and-Language Pre-training towards Unified Retrieval and Captioning
15:30-17:00
End-to-End Unsupervised Vision-and-Language Pre-training with Referring Expression Matching
15:30-17:00
Retrieval Augmented Visual Question Answering with Outside Knowledge
15:30-17:00
Few-shot Query-Focused Summarization with Prefix-Merging
15:30-17:00
R-TeaFor: Regularized Teacher-Forcing for Abstractive Summarization
15:30-17:00
Towards Summary Candidates Fusion
15:30-17:00
HEGEL: Hypergraph Transformer for Long Document Summarization
15:30-17:00
CiteSum: Citation Text-guided Scientific Extreme Summarization and Domain Adaptation with Limited Supervision
15:30-17:00
Unsupervised Tokenization Learning
15:30-17:00
FastClass: A Time-Efficient Approach to Weakly-Supervised Text Classification
15:30-17:00
CycleKQR: Unsupervised Bidirectional Keyword-Question Rewriting
15:30-17:00
STRUDEL: Structured Dialogue Summarization for Dialogue Comprehension
15:30-17:00
CONQRR: Conversational Query Rewriting for Retrieval with Reinforcement Learning
15:30-17:00
DialogConv: A Lightweight Fully Convolutional Network for Multi-view Response Selection
15:30-17:00
Hardness-guided domain adaptation to recognise biomedical named entities under low-resource scenarios
15:30-17:00
Overcoming Catastrophic Forgetting in Zero-Shot Cross-Lingual Generation
15:30-17:00
Logical Neural Networks for Knowledge Base Completion with Embeddings & Rules
15:30-17:00
Topic Modeling With Topological Data Analysis
15:30-17:00
Measuring the Mixing of Contextual Information in the Transformer
15:30-17:00
Nearest Neighbor Zero-Shot Inference
15:30-17:00
Fine-tuned Language Models are Continual Learners
15:30-17:00
Boosting Natural Language Generation from Instructions with Meta-Learning
15:30-17:00
Passage-Mask: A Learnable Regularization Strategy for Retriever-Reader Models
15:30-17:00
Fixing Model Bugs with Natural Language Patches
15:30-17:00
GREENER: Graph Neural Networks for News Media Profiling
15:30-17:00
Self-supervised Graph Masking Pre-training for Graph-to-Text Generation
15:30-17:00
On the Evaluation Metrics for Paraphrase Generation
ExPUNations: Augmenting Puns with Keywords and Explanations
15:30-17:00
RobustLR: A Diagnostic Benchmark for Evaluating Logical Robustness of Deductive Reasoners
15:30-17:00
SCROLLS: Standardized CompaRison Over Long Language Sequences
15:30-17:00
Semantic-aware Contrastive Learning for More Accurate Semantic Parsing
15:30-17:00
Algorithms for Weighted Pushdown Automata
15:30-17:00
[CL] Nucleus Composition in Transition-Based Dependency Parsing
15:30-17:00
A Major Obstacle for NLP Research: Let’s Talk about Time Allocation!
15:30-17:00
Unsupervised Entity Linking with Guided Summarization and Multiple-Choice Selection