Timezone: Conference (Toronto) UTC Browser
⏰ Time: July 09, 09:00-09:00 (EDT) Conference
⏰ Time: (13:00-13:00 UTC)
UTC /
Browser
⏰ Time: July 09, 09:00-09:00 (EDT) Conference
⏰ Time: (13:00-13:00 UTC)
UTC /
Browser
⏰ Time: July 09, 09:00-09:00 (EDT) Conference
⏰ Time: (13:00-13:00 UTC)
UTC /
Browser
⏰ Time: July 09, 14:00-14:00 (EDT) Conference
⏰ Time: (18:00-18:00 UTC)
UTC /
Browser
⏰ Time: July 09, 14:00-14:00 (EDT) Conference
⏰ Time: (18:00-18:00 UTC)
UTC /
Browser
⏰ Time: July 09, 14:00-14:00 (EDT) Conference
⏰ Time: (18:00-18:00 UTC)
UTC /
Browser
Timezone: Conference (Toronto) UTC Browser
Demo Session 1
⏰ Time: July 10, 11:00-12:30 (EDT) Conference
⏰ Time: (15:00-16:30 UTC)
UTC /
Browser
Poster Presentations
Generation (demo) (Poster)
Room: Frontenac Ballroom and Queen's Quay
- [Demo] Disco: a toolkit for Distributional Control of Generative Models
- [Demo] SciLit: A Platform for Joint Scientific Literature Discovery, Summarization and Citation Generation
- [Demo] CB2: Collaborative Natural Language Interaction Research Platform
- [Demo] Fast Whitespace Correction with Encoder-Only Transformers
Large Language Models (demo) (Poster)
Room: Frontenac Ballroom and Queen's Quay
Speech and Multimodality (demo) (Poster)
Room: Frontenac Ballroom and Queen's Quay
Demo Session 2
⏰ Time: July 10, 14:00-15:30 (EDT) Conference
⏰ Time: (18:00-19:30 UTC)
UTC /
Browser
Poster Presentations
Dialogue and Interactive Systems (demo) (Poster)
Room: Frontenac Ballroom and Queen's Quay
Generation (demo) (Poster)
Room: Frontenac Ballroom and Queen's Quay
Information Extraction (demo) (Poster)
Room: Frontenac Ballroom and Queen's Quay
Multilingualism and Cross-Lingual NLP (demo) (Poster)
Room: Frontenac Ballroom and Queen's Quay
Speech and Multimodality (demo) (Poster)
Room: Frontenac Ballroom and Queen's Quay
Poster Session 1
⏰ Time: July 10, 11:00-12:30 (EDT) Conference
⏰ Time: (15:00-16:30 UTC)
UTC /
Browser
Poster Presentations
Computational Social Science and Cultural Analytics (Poster)
Room: Frontenac Ballroom and Queen's Quay
Dialogue and Interactive Systems (Poster)
Room: Frontenac Ballroom and Queen's Quay
Ethics and NLP (Poster)
Room: Frontenac Ballroom and Queen's Quay
Generation (Poster)
Room: Frontenac Ballroom and Queen's Quay
Information Extraction (Poster)
Room: Frontenac Ballroom and Queen's Quay
- Causality-aware Concept Extraction based on Knowledge-guided Prompting
- Information Screening whilst Exploiting! Multimodal Relation Extraction with Feature Denoising and Multimodal Topic Modeling
- DiffusionNER: Boundary Diffusion for Named Entity Recognition
- Modeling Instance Interactions for Joint Information Extraction with Neural High-Order Conditional Random Field
- Split-NER: Named Entity Recognition via Two Question-Answering-based Classifications
- RE-Matching: A Fine-Grained Semantic Matching Method for Zero-Shot Relation Extraction
- Few-Shot Document-Level Event Argument Extraction
- Few-shot Event Detection: An Empirical Study and a Unified View
- Simple Augmentations of Logical Rules for Neuro-Symbolic Knowledge Graph Completion
- Continual Contrastive Finetuning Improves Low-Resource Relation Extraction
Information Retrieval and Text Mining (Poster)
Room: Frontenac Ballroom and Queen's Quay
Interpretability and Analysis of Models for NLP (Poster)
Room: Frontenac Ballroom and Queen's Quay
Language Grounding to Vision, Robotics, and Beyond (Poster)
Room: Frontenac Ballroom and Queen's Quay
Large Language Models (Poster)
Room: Frontenac Ballroom and Queen's Quay
- Training-free Neural Architecture Search for RNNs and Transformers
- Rethinking the Role of Scale for In-Context Learning: An Interpretability-based Case Study at 66 Billion Scale
- In-Context Analogical Reasoning with Pre-Trained Language Models
- Symbolic Chain-of-Thought Distillation: Small Models Can Also "Think" Step-by-Step
- Did You Read the Instructions? Rethinking the Effectiveness of Task Definitions in Instruction Learning
Linguistic Theories, Cognitive Modeling, and Psycholinguistics (Poster)
Room: Frontenac Ballroom and Queen's Quay
Machine Learning for NLP (Poster)
Room: Frontenac Ballroom and Queen's Quay
- Downstream Datasets Make Surprisingly Good Pretraining Corpora
- miCSE: Mutual Information Contrastive Learning for Low-shot Sentence Embeddings
- Randomized Positional Encodings Boost Length Generalization of Transformers
- Large-scale Lifelong Learning of In-context Instructions and How to Tackle It
- HyperMixer: An MLP-based Low Cost Alternative to Transformers
- HuCurl: Human-induced Curriculum Discovery
Machine Translation (Poster)
Room: Frontenac Ballroom and Queen's Quay
Multilingualism and Cross-Lingual NLP (Poster)
Room: Frontenac Ballroom and Queen's Quay
- Crosslingual Generalization through Multitask Finetuning
- mPMR: A Multilingual Pre-trained Machine Reader at Scale
- BLOOM+1: Adding Language Support to BLOOM for Zero-Shot Prompting
- Beyond Contrastive Learning: A Variational Generative Model for Multilingual Retrieval
- Soft Language Clustering for Multilingual Model Pre-training
NLP Applications (Poster)
Room: Frontenac Ballroom and Queen's Quay
Question Answering (Poster)
Room: Frontenac Ballroom and Queen's Quay
Resources and Evaluation (Poster)
Room: Frontenac Ballroom and Queen's Quay
- WikiBio: a Semantic Resource for the Intersectional Analysis of Biographical Events
- Bhasa-Abhijnaanam: Native-script and romanized Language Identification for 22 Indic languages
- EPIC: Multi-Perspective Annotation of a Corpus of Irony
- Do language models have coherent mental models of everyday things?
- NLPeer: A Unified Resource for the Computational Study of Peer Review
- Multilingual Multifaceted Understanding of Online News in Terms of Genre, Framing, and Persuasion Techniques
Semantics: Lexical (Poster)
Room: Frontenac Ballroom and Queen's Quay
Semantics: Sentence-level Semantics, Textual Inference, and Other Areas (Poster)
Room: Frontenac Ballroom and Queen's Quay
- Going Beyond Sentence Embeddings: A Token-Level Matching Algorithm for Calculating Semantic Textual Similarity
- APOLLO: A Simple Approach for Adaptive Pretraining of Language Models for Logical Reasoning
- Evaluating Paraphrastic Robustness in Textual Entailment Models
- Label-Aware Hyperbolic Embeddings for Fine-grained Emotion Classification
- Learning Action Conditions from Instructional Manuals for Instruction Understanding
Sentiment Analysis, Stylistic Analysis, and Argument Mining (Poster)
Room: Frontenac Ballroom and Queen's Quay
Speech and Multimodality (Poster)
Room: Frontenac Ballroom and Queen's Quay
Summarization (Poster)
Room: Frontenac Ballroom and Queen's Quay
- DIONYSUS: A Pre-trained Model for Low-Resource Dialogue Summarization
- Generating EDU Extracts for Plan-Guided Summary Re-Ranking
- Improving the Robustness of Summarization Systems with Dual Augmentation
- Peek Across: Improving Multi-Document Modeling via Cross-Document Question-Answering
- Toward Expanding the Scope of Radiology Report Summarization to Multiple Anatomies and Modalities
- Automated Metrics for Medical Multi-Document Summarization Disagree with Human Evaluations
- Concise Answers to Complex Questions: Summarization of Long-form Answers
- Extractive is not Faithful: An Investigation of Broad Unfaithfulness Problems in Extractive Summarization
Syntax: Tagging, Chunking, and Parsing (Poster)
Room: Frontenac Ballroom and Queen's Quay
Theme: Reality Check (Poster)
Room: Frontenac Ballroom and Queen's Quay
Poster Session 2
⏰ Time: July 10, 14:00-15:30 (EDT) Conference
⏰ Time: (18:00-19:30 UTC)
UTC /
Browser
Poster Presentations
Computational Social Science and Cultural Analytics (Poster)
Room: Frontenac Ballroom and Queen's Quay
Dialogue and Interactive Systems (Poster)
Room: Frontenac Ballroom and Queen's Quay
Discourse and Pragmatics (Poster)
Room: Frontenac Ballroom and Queen's Quay
Ethics and NLP (Poster)
Room: Frontenac Ballroom and Queen's Quay
Generation (Poster)
Room: Frontenac Ballroom and Queen's Quay
Information Extraction (Poster)
Room: Frontenac Ballroom and Queen's Quay
- More than Classification: A Unified Framework for Event Temporal Relation Extraction
- Discriminative Reasoning with Sparse Event Representation for Document-level Event-Event Relation Extraction
- Rethinking Multimodal Entity and Relation Extraction from a Translation Point of View
- When to Use What: An In-Depth Comparative Empirical Analysis of OpenIE Systems for Downstream Applications
Interpretability and Analysis of Models for NLP (Poster)
Room: Frontenac Ballroom and Queen's Quay
Language Grounding to Vision, Robotics, and Beyond (Poster)
Room: Frontenac Ballroom and Queen's Quay
Large Language Models (Poster)
Room: Frontenac Ballroom and Queen's Quay
- Verify-and-Edit: A Knowledge-Enhanced Chain-of-Thought Framework
- Latent Positional Information is in the Self-Attention Variance of Transformer Language Models Without Positional Embeddings
- Few-shot In-context Learning on Knowledge Base Question Answering
- MixCE: Training Autoregressive Language Models by Mixing Forward and Reverse Cross-Entropies
- Do Models Really Learn to Follow Instructions? An Empirical Study of Instruction Tuning
- Say What You Mean! Large Language Models Speak Too Positively about Negative Commonsense Knowledge
- Mitigating Label Biases for In-context Learning
- Open-Domain Hierarchical Event Schema Induction by Incremental Prompting and Verification
- RL4F: Generating Natural Language Feedback with Reinforcement Learning for Repairing Model Outputs
- Training Trajectories of Language Models Across Scales
- NarrowBERT: Accelerating Masked Language Model Pretraining and Inference
Linguistic Theories, Cognitive Modeling, and Psycholinguistics (Poster)
Room: Frontenac Ballroom and Queen's Quay
Machine Learning for NLP (Poster)
Room: Frontenac Ballroom and Queen's Quay
- ColD Fusion: Collaborative Descent for Distributed Multitask Finetuning
- CELDA: Leveraging Black-box Language Model as Enhanced Classifier without Labels
- Tailoring Instructions to Student's Learning Levels Boosts Knowledge Distillation
- Backpack Language Models
- Targeted Data Generation: Finding and Fixing Model Weaknesses
Machine Translation (Poster)
Room: Frontenac Ballroom and Queen's Quay
- Accelerating Transformer Inference for Translation via Parallel Decoding
- The Inside Story: Towards Better Understanding of Machine Translation Neural Evaluation Metrics
- A Simple Concatenation can Effectively Improve Speech Translation
- INK: Injecting kNN Knowledge in Nearest Neighbor Machine Translation
- Local Byte Fusion for Neural Machine Translation
Multilingualism and Cross-Lingual NLP (Poster)
Room: Frontenac Ballroom and Queen's Quay
NLP Applications (Poster)
Room: Frontenac Ballroom and Queen's Quay
- Zero-shot Faithful Factual Error Correction
- Improving Automatic Quotation Attribution in Literary Novels
- Improving Domain Generalization for Prompt-Aware Essay Scoring via Disentangled Representation Learning
- LeXFiles and LegalLAMA: Facilitating English Multinational Legal Language Model Development
- Interpretable Math Word Problem Solution Generation via Step-by-step Planning
- Vision Meets Definitions: Unsupervised Visual Word Sense Disambiguation Incorporating Gloss Information
- Logic-driven Indirect Supervision: An Application to Crisis Counseling
- BIC: Twitter Bot Detection with Text-Graph Interaction and Semantic Consistency
Phonology, Morphology, and Word Segmentation (Poster)
Room: Frontenac Ballroom and Queen's Quay
Question Answering (Poster)
Room: Frontenac Ballroom and Queen's Quay
- Task-Aware Specialization for Efficient and Robust Dense Retrieval for Open-Domain Question Answering
- MultiTool-CoT: GPT-3 Can Use Multiple External Tools with Chain of Thought Prompting
- Learning to Simulate Natural Language Feedback for Interactive Semantic Parsing
- BUCA: A Binary Classification Approach to Unsupervised Commonsense Question Answering
- Answering Ambiguous Questions via Iterative Prompting
- DisentQA: Disentangling Parametric and Contextual Knowledge with Counterfactual Question Answering
- Using contradictions improves question answering systems
- Elaboration-Generating Commonsense Question Answering at Scale
- MatCha: Enhancing Visual Language Pretraining with Math Reasoning and Chart Derendering
Resources and Evaluation (Poster)
Room: Frontenac Ballroom and Queen's Quay
- A Survey of Deep Learning for Mathematical Reasoning
- IDRISI-RA: The First Arabic Location Mention Recognition Dataset of Disaster Tweets
- Exploring Large Language Models for Classical Philology
- The KITMUS Test: Evaluating Knowledge Integration from Multiple Sources
- Revisiting non-English Text Simplification: A Unified Multilingual Benchmark
- A Needle in a Haystack: An Analysis of High-Agreement Workers on MTurk for Summarization
- DEplain: A German Parallel Corpus with Intralingual Translations into Plain Language for Sentence and Document Simplification
- Naamapadam: A Large-Scale Named Entity Annotated Data for Indic Languages
- MDACE: MIMIC Documents Annotated with Code Evidence
Semantics: Lexical (Poster)
Room: Frontenac Ballroom and Queen's Quay
Semantics: Sentence-level Semantics, Textual Inference, and Other Areas (Poster)
Room: Frontenac Ballroom and Queen's Quay
Sentiment Analysis, Stylistic Analysis, and Argument Mining (Poster)
Room: Frontenac Ballroom and Queen's Quay
Speech and Multimodality (Poster)
Room: Frontenac Ballroom and Queen's Quay
Summarization (Poster)
Room: Frontenac Ballroom and Queen's Quay
- Annotating and Detecting Fine-grained Factual Errors for Dialogue Summarization
- Multi-Document Summarization with Centroid-Based Pretraining
- Abstractive Summarizers are Excellent Extractive Summarizers
- On Improving Summarization Factual Consistency from Natural Language Feedback
- Towards Understanding Omission in Dialogue Summarization
Syntax: Tagging, Chunking, and Parsing (Poster)
Room: Frontenac Ballroom and Queen's Quay
Theme: Reality Check (Poster)
Room: Frontenac Ballroom and Queen's Quay
Session 1
⏰ Time: July 10, 11:00-12:30 (EDT) Conference
⏰ Time: (15:00-16:30 UTC)
UTC /
Browser
Oral Presentations
Ethics and NLP (Oral)
Room: Pier 2&3
- [TACL] Hate Speech Classifiers Learn Normative Social Stereotypes
- NLPositionality: Characterizing Design Biases of Datasets and Models
- What social attitudes about gender does BERT encode? Leveraging insights from psycholinguistics
- The Elephant in the Room: Analyzing the Presence of Big Tech in Natural Language Processing Research
- WinoQueer: A Community-in-the-Loop Benchmark for Anti-LGBTQ+ Bias in Large Language Models
- ETHICIST: Targeted Training Data Extraction Through Loss Smoothed Soft Prompting and Calibrated Confidence Estimation
Large Language Models (Oral)
Room: Metropolitan Centre
- Searching for Needles in a Haystack: On the Role of Incidental Bilingualism in PaLM's Translation Capability
- KILM: Knowledge Injection into Encoder-Decoder Language Models
- When Not to Trust Language Models: Investigating Effectiveness of Parametric and Non-Parametric Memories
- Unified Demonstration Retriever for In-Context Learning
- Measuring Inductive Biases of In-Context Learning with Underspecified Demonstrations
- Prompting PaLM for Translation: Assessing Strategies and Performance
Multilingualism and Cross-Lingual NLP (Oral)
Room: Pier 4&5
- Improving the Detection of Multilingual Online Attacks with Rich Social Media Data from Singapore
- MASSIVE: A 1M-Example Multilingual Natural Language Understanding Dataset with 51 Typologically-Diverse Languages
- Empowering Cross-lingual Behavioral Testing of NLP Models with Typological Features
- [CL] Data-driven Cross-lingual Syntax: An Agreement Study with Massively Multilingual Models
- Towards Zero-Shot Multilingual Transfer for Code-Switched Responses
- On Evaluating Multilingual Compositional Generalization with Translated Datasets
NLP Applications (Oral)
Room: Metropolitan East
- CoAD: Automatic Diagnosis through Symptom and Disease Collaborative Generation
- Clinical Note Owns its Hierarchy: Multi-Level Hypergraph Neural Networks for Patient-Level Representation Learning
- DICE: Data-Efficient Clinical Event Extraction with Generative Models
- TemplateGEC: Improving Grammatical Error Correction with Detection Template
- Towards Domain-Agnostic and Domain-Adaptive Dementia Detection from Spoken Language
- Using Neural Machine Translation for Generating Diverse Challenging Exercises for Language Learner
Question Answering (Oral)
Room: Metropolitan West
- Multi-Source Test-Time Adaptation as Dueling Bandits for Extractive Question Answering
- Query Refinement Prompts for Closed-Book Long-Form QA
- Won't Get Fooled Again: Answering Questions with False Premises
- Abductive Commonsense Reasoning Exploiting Mutually Exclusive Explanations
- To Adapt or to Annotate: Challenges and Interventions for Domain Adaptation in Open-Domain Question Answering
- Post-Abstention: Towards Reliably Re-Attempting the Abstained Instances in QA
Poster Presentations
Dialogue and Interactive Systems (Poster)
Room: Frontenac Ballroom and Queen's Quay
Generation (Poster)
Room: Frontenac Ballroom and Queen's Quay
Machine Learning for NLP (Poster)
Room: Frontenac Ballroom and Queen's Quay
Semantics: Sentence-level Semantics, Textual Inference, and Other Areas (Poster)
Room: Frontenac Ballroom and Queen's Quay
Sentiment Analysis, Stylistic Analysis, and Argument Mining (Poster)
Room: Frontenac Ballroom and Queen's Quay
Virtual Poster Presentations
Computational Social Science and Cultural Analytics (Virtual Poster)
Room: Pier 7&8
- NormBank: A Knowledge Bank of Situational Social Norms
- Counterfactual Probing for the Influence of Affect and Specificity on Intergroup Bias
- Modeling Cross-Cultural Pragmatic Inference with Codenames Duet
- Interactive Concept Learning for Uncovering Latent Themes in Large Text Collections
- Grounded Multimodal Named Entity Recognition on Social Media
- UPPAM: A Unified Pre-training Architecture for Political Actor Modeling based on Language
- Additive manifesto decomposition: A policy domain aware method for understanding party positioning
- COBRA Frames: Contextual Reasoning about Effects and Harms of Offensive Statements
- MultiEMO: An Attention-Based Correlation-Aware Multimodal Fusion Framework for Emotion Recognition in Conversations
Dialogue and Interactive Systems (Virtual Poster)
Room: Pier 7&8
- Query Enhanced Knowledge-Intensive Conversation via Unsupervised Joint Modeling
- Boosting Distress Support Dialogue Responses with Motivational Interviewing Strategy
- Leveraging Explicit Procedural Instructions for Data-Efficient Action Prediction
- CASE: Aligning Coarse-to-Fine Cognition and Affection for Empathetic Response Generation
- End-to-End Task-Oriented Dialogue Systems Based on Schema
- Ask an Expert: Leveraging Language Models to Improve Strategic Reasoning in Goal-Oriented Dialogue Models
- NewsDialogues: Towards Proactive News Grounded Conversation
- PaCE: Unified Multi-modal Dialogue Pre-training with Progressive and Compositional Experts
- Speech-Text Pre-training for Spoken Dialog Understanding with Explicit Cross-Modal Alignment
- Robust Learning for Multi-party Addressee Recognition with Discrete Addressee Codebook
- CausalDialogue: Modeling Utterance-level Causality in Conversations
- Intent Discovery with Frame-guided Semantic Regularization and Augmentation
- DEnsity: Open-domain Dialogue Evaluation Metric using Density Estimation
- Retrieval-free Knowledge Injection through Multi-Document Traversal for Dialogue Models
- Multi3NLU++: A Multilingual, Multi-Intent, Multi-Domain Dataset for Natural Language Understanding in Task-Oriented Dialogue
- Prompted LLMs as Chatbot Modules for Long Open-domain Conversation
- Zero-Shot Prompting for Implicit Intent Prediction and Recommendation with Commonsense Reasoning
- CORE: Cooperative Training of Retriever-Reranker for Effective Dialogue Response Selection
- Diverse Retrieval-Augmented In-Context Learning for Dialogue State Tracking
- One Cannot Stand for Everyone! Leveraging Multiple User Simulators\\ to train Task-oriented Dialogue Systems
- Towards Fewer Hallucinations in Knowledge-Grounded Dialogue Generation via Augmentative and Contrastive Knowledge-Dialogue
- EM Pre-training for Multi-party Dialogue Response Generation
- Medical Dialogue Generation via Dual Flow Modeling
- Towards Open Environment Intent Prediction
- A Synthetic Data Generation Framework for Grounded Dialogues
Discourse and Pragmatics (Virtual Poster)
Room: Pier 7&8
- How Well Do Large Language Models Perform on Faux Pas Tests?
- PragmatiCQA: A Dataset for Pragmatic Question Answering in Conversations
- A Match Made in Heaven: A Multi-task Framework for Hyperbole and Metaphor Detection
- The Coreference under Transformation Labeling Dataset: Entity Tracking in Procedural Texts Using Event Models
- Learning Event-aware Measures for Event Coreference Resolution
- $2*n$ is better than $n^2$: Decomposing Event Coreference Resolution into Two Tractable Problems
Ethics and NLP (Virtual Poster)
Room: Pier 7&8
- Bias Beyond English: Counterfactual Tests for Bias in Sentiment Analysis in Four Languages
- A Multi-dimensional study on Bias in Vision-Language models
- D-CALM: A Dynamic Clustering-based Active Learning Approach for Mitigating Bias
- Causal-Debias: Unifying Debiasing in Pretrained Language Models and Fine-tuning via Causal Invariant Learning
- Uncovering and Categorizing Social Biases in Text-to-SQL
- Nichelle and Nancy: The Influence of Demographic Attributes and Tokenization Length on First Name Biases
- With Prejudice to None: A Few-Shot, Multilingual Transfer Learning Approach to Detect Social Bias in Low Resource Languages
Generation (Virtual Poster)
Room: Pier 7&8
- [TACL] Directed Acyclic Transformer Pre-training for High-quality Non-autoregressive Text Generation
- UniLG: A Unified Structure-aware Framework for Lyrics Generation
- Critic-Guided Decoding for Controlled Text Generation
- Controllable Text Generation via Probability Density Estimation in the Latent Space
- TAVT: Towards Transferable Audio-Visual Text Generation
- RARR: Researching and Revising What Language Models Say, Using Language Models
- Revisiting Sentence Union Generation as a Testbed for Text Consolidation
- LLM-Blender: Ensembling Large Language Models with Pairwise Ranking and Generative Fusion
- Explicit Syntactic Guidance for Neural Text Generation
- Language Modeling with Latent Situations
- Evaluation of Question Generation Needs More References
- Open-ended Long Text Generation via Masked Language Modeling
- Focused Prefix Tuning for Controllable Text Generation
- Controlled Text Generation with Hidden Representation Transformations
- A New Dataset and Empirical Study for Sentence Simplification in Chinese
Information Extraction (Virtual Poster)
Room: Pier 7&8
- A Class-Rebalancing Self-Training Framework for Distantly-Supervised Named Entity Recognition
- Revisiting Event Argument Extraction: Can EAE Models Learn Better When Being Aware of Event Co-occurrences?
- PromptRank: Unsupervised Keyphrase Extraction Using Prompt
- LayoutMask: Enhance Text-Layout Interaction in Multi-modal Pre-training for Document Understanding
- Joint Constrained Learning with Boundary-adjusting for Emotion-Cause Pair Extraction
- UniEX: An Effective and Efficient Framework for Unified Information Extraction via a Span-extractive Perspective
- Enhancing Event Causality Identification with Event Causal Label and Event Pair Interaction Graph
- A Diffusion Model for Event Skeleton Generation
- Towards Better Entity Linking with Multi-View Enhanced Distillation
- Benchmarking Diverse-Modal Entity Linking with Generative Models
- Concept2Box: Joint Geometric Embeddings for Learning Two-View Knowledge Graphs
- Understanding Demonstration-based Learning from a Causal Perspective
- FSUIE: A Novel Fuzzy Span Mechanism for Universal Information Extraction
- Learning In-context Learning for Named Entity Recognition
- Class-Adaptive Self-Training for Relation Extraction with Incompletely Annotated Training Data
- Double-Branch Multi-Attention based Graph Neural Network for Knowledge Graph Completion
- Recall, Expand, and Multi-Candidate Cross-Encode: Fast and Accurate Ultra-Fine Entity Typing
Information Retrieval and Text Mining (Virtual Poster)
Room: Pier 7&8
- Structure-Aware Language Model Pretraining Improves Dense Retrieval on Structured Data
- Enhancing Hierarchical Text Classification through Knowledge Graph Integration
- Discrete Prompt Optimization via Constrained Generation for Zero-shot Re-ranker
- UniTRec: A Unified Text-to-Text Transformer and Joint Contrastive Learning Framework for Text-based Recommendation
- Dynamic Structured Neural Topic Model with Self-Attention Mechanism
- Boosting Zero-shot Cross-lingual Retrieval by Training on Artificially Code-Switched Data
Interpretability and Analysis of Models for NLP (Virtual Poster)
Room: Pier 7&8
- Inducing Character-level Structure in Subword-based Language Models with Type-level Interchange Intervention Training
- Limitations of Language Models in Arithmetic and Symbolic Induction
- Explanation Regeneration via Information Bottleneck
- Characterizing the Impacts of Instances on Robustness
- COCKATIEL: COntinuous Concept ranKed ATtribution with Interpretable ELements for explaining neural net classifiers on NLP
- Interpreting Positional Information in Perspective of Word Order
- Model Interpretability and Rationale Extraction by Input Mask Optimization
- Nonlinear Structural Equation Model Guided Gaussian Mixture Hierarchical Topic Modeling
- Layerwise universal adversarial attack on NLP models
- Robust Natural Language Understanding with Residual Attention Debiasing
- Deep Model Compression Also Helps Models Capture Ambiguity
- On Prefix-tuning for Lightweight Out-of-distribution Detection
- Towards Stable Natural Language Understanding via Information Entropy Guided Debiasing
- Transformer Language Models Handle Word Frequency in Prediction Head
- Is Continuous Prompt a Combination of Discrete Prompts? Towards a Novel View for Interpreting Continuous Prompts
- Causal interventions expose implicit situation models for commonsense language understanding
- Constructing Word-Context-Coupled Space Aligned with Associative Knowledge Relations for Interpretable Language Modeling
- Easy to Decide, Hard to Agree: Reducing Disagreements Between Saliency Methods
- HyHTM: Hyperbolic Geometry-based Hierarchical Topic Model
- Do PLMs Know and Understand Ontological Knowledge?
Language Grounding to Vision, Robotics, and Beyond (Virtual Poster)
Room: Pier 7&8
- CONE: An Efficient COarse-to-fiNE Alignment Framework for Long Video Temporal Grounding
- Cross-View Language Modeling: Towards Unified Cross-Lingual Cross-Modal Pre-training
- Multimedia Generative Script Learning for Task Planning
- Translation-Enhanced Multilingual Text-to-Image Generation
- PV2TEA: Patching Visual Modality to Textual-Established Information Extraction
- A Neural Divide-and-Conquer Reasoning Framework for Image Retrieval from Linguistically Complex Text
- Enhanced Chart Understanding via Visual Language Pre-training on Plot Table Pairs
- Towards Parameter-Efficient Integration of Pre-Trained Language Models In Temporal Video Grounding
- Evaluating pragmatic abilities of image captioners on A3DS
- Visual Coherence Loss for Coherent and Visually Grounded Story Generation
- Define, Evaluate, and Improve Task-Oriented Cognitive Capabilities for Instruction Generation Models
- AltCLIP: Altering the Language Encoder in CLIP for Extended Language Capabilities
- Segment-Level and Category-Oriented Network for Knowledge-Based Referring Expression Comprehension
- Speaking the Language of Your Listener: Audience-Aware Adaptation via Plug-and-Play Theory of Mind
Language Grounding to Vision, Robotics and Beyond (demo) (Virtual Poster)
Room: Pier 7&8
Large Language Models (Virtual Poster)
Room: Pier 7&8
- Code Execution with Pre-trained Language Models
- HiFi: High-Information Attention Heads Hold for Parameter-Efficient Model Adaptation
- How does the task complexity of masked pretraining objectives affect downstream performance?
- ThinkSum: Probabilistic reasoning over sets using large language models
- Hierarchical Verbalizer for Few-Shot Hierarchical Text Classification
- Recyclable Tuning for Continual Pre-training
- Contrastive Novelty-Augmented Learning: Anticipating Outliers with Large Language Models
- The Magic of IF: Investigating Causal Reasoning Abilities in Large Language Models of Code
- Membership Inference Attacks against Language Models via Neighbourhood Comparison
- Complementary Explanations for Effective In-Context Learning
- Aligning Instruction Tasks Unlocks Large Language Models as Zero-Shot Relation Extractors
- Model-Generated Pretraining Signals Improves Zero-Shot Generalization of Text-to-Text Transformers
- MultiInstruct: Improving Multi-Modal Zero-Shot Learning via Instruction Tuning
- Unnatural Instructions: Tuning Language Models with (Almost) No Human Labor
- Making Language Models Better Reasoners with Step-Aware Verifier
- Black-box language model explanation by context length probing
- Revisiting Token Dropping Strategy in Efficient BERT Pretraining
- Attribute Controlled Dialogue Prompting
- Farewell to Aimless Large-scale Pretraining: Influential Subset Selection for Language Model
- Recursion of Thought: A Divide-and-Conquer Approach to Multi-Context Reasoning with Language Models
- Learning Better Masking for Better Language Model Pre-training
Linguistic Theories, Cognitive Modeling, and Psycholinguistics (Virtual Poster)
Room: Pier 7&8
Machine Learning for NLP (demo) (Virtual Poster)
Room: Pier 7&8
Machine Learning for NLP (Virtual Poster)
Room: Pier 7&8
- [TACL] Minimum Description Length Recurrent Neural Networks
- Scale-Invariant Infinite Hierarchical Topic Model
- Zero-Shot Classification by Logical Reasoning on Natural Language Explanations
- EmbedTextNet: Dimension Reduction with Weighted Reconstruction and Correlation Losses for Efficient Text Embedding
- A Memory Model for Question Answering from Streaming Data Supported by Rehearsal and Anticipation of Coreference Information
- Class-Incremental Learning based on Label Generation
- Enhancing Out-of-Vocabulary Estimation with Subword Attention
- PreQuant: A Task-agnostic Quantization Approach for Pre-trained Language Models
- Domain Aligned Prefix Averaging for Domain Generalization in Abstractive Summarization
- Not Enough Data to Pre-train Your Language Model? MT to the Rescue!
- Label Agnostic Pre-training for Zero-shot Text Classification
- MANNER: A Variational Memory-Augmented Model for Cross Domain Few-Shot Named Entity Recognition
- TeAST: Temporal Knowledge Graph Embedding via Archimedean Spiral Timeline
- DSEE: Dually Sparsity-embedded Efficient Tuning of Pre-trained Language Models
- Text Adversarial Purification as Defense against Adversarial Attacks
- TART: Improved Few-shot Text Classification Using Task-Adaptive Reference Transformation
- DaMSTF: Domain Adversarial Learning Enhanced Meta Self-Training for Domain Adaptation
- AutoMoE: Heterogeneous Mixture-of-Experts with Adaptive Computation for Efficient Neural Machine Translation
- Neural Architecture Search for Parameter-Efficient Fine-tuning of Large Pre-trained Language Models
- Simplicity Bias in Transformers and their Ability to Learn Sparse Boolean Functions
- Structured Mean-Field Variational Inference for Higher-Order Span-Based Semantic Role
- Towards Distribution-shift Robust Text Classification of Emotional Content
- Boosting Text Augmentation via Hybrid Instance Filtering Framework
Machine Translation (Virtual Poster)
Room: Pier 7&8
- Bridging the Domain Gaps in Context Representations for $k$-Nearest Neighbor Neural Machine Translation
- Understanding and Improving the Robustness of Terminology Constraints in Neural Machine Translation
- Target-Side Augmentation for Document-Level Machine Translation
- Towards Speech Dialogue Translation Mediating Speakers of Different Languages
- DUB: Discrete Unit Back-translation for Speech Translation
- Synthetic Pre-Training Tasks for Neural Machine Translation
- Do GPTs Produce Less Literal Translations?
- Back Translation for Speech-to-text Translation Without Transcripts
- Understanding and Bridging the Modality Gap for Speech Translation
- Rethinking the Word-level Quality Estimation for Machine Translation from Human Judgement
- A Little is Enough: Few-Shot Quality Estimation based Corpus Filtering improves Machine Translation
- A Holistic Approach to Reference-Free Evaluation of Machine Translation
Multilingualism and Cross-Lingual NLP (Virtual Poster)
Room: Pier 7&8
- Automatic Identification of Code-Switching Functions in Speech Transcripts
- Exploring the Relationship between Alignment and Cross-lingual Transfer in Multilingual Transformers
- Typology Guided Multilingual Position Representations: Case on Dependency Parsing
- Speaking Multiple Languages Affects the Moral Bias of Language Models
- Dual-Alignment Pre-training for Cross-lingual Sentence Embedding
- TADA : Task Agnostic Dialect Adapters for English
NLP Applications (demo) (Virtual Poster)
Room: Pier 7&8
NLP Applications (Virtual Poster)
Room: Pier 7&8
- Cross-lingual Science Journalism: Select, Simplify and Rewrite Summaries for Non-expert Readers
- A Two-Stage Decoder for Efficient ICD Coding
- Score It All Together: A Multi-Task Learning Study on Automatic Scoring of Argumentative Essays
- Wukong-Reader: Multi-modal Pre-training for Fine-grained Visual Document Understanding
- FEDLEGAL: The First Real-World Federated Learning Benchmark for Legal NLP
- Replace and Report: NLP Assisted Radiology Report Generation
- Similarity-Based Content Scoring - A more Classroom-Suitable Alternative to Instance-Based Scoring?
- The Mechanical Bard: An Interpretable Machine Learning Approach to Shakespearean Sonnet Generation
- Contrastive Learning with Generated Representations for Inductive Knowledge Graph Embedding
- Counterfactual Debiasing for Fact Verification
- Hard Sample Aware Prompt-Tuning
- RMLM: A Flexible Defense Framework for Proactively Mitigating Word-level Adversarial Attacks
- Understanding Programs by Exploiting (Fuzzing) Test Cases
- Distantly Supervised Course Concept Extraction in MOOCs with Academic Discipline
- Dynamic Routing Transformer Network for Multimodal Sarcasm Detection
- TECHS: Temporal Logical Graph Networks for Explainable Extrapolation Reasoning
- UniEvent: Unified Generative Model with Multi-Dimensional Prefix for Zero-Shot Event-Relational Reasoning
- Learning Query Adaptive Anchor Representation for Inductive Relation Prediction
- ClaimDiff: Comparing and Contrasting Claims on Contentious Issues
Question Answering (Virtual Poster)
Room: Pier 7&8
- Faithful Question Answering with Monte-Carlo Planning
- Expand, Rerank, and Retrieve: Query Reranking for Open-Domain Question Answering
- An Empirical Comparison of LM-based Question and Answer Generation Methods
- Combo of Thinking and Observing for Outside-Knowledge VQA
- Chain-of-Skills: A Configurable Model for Open-Domain Question Answering
- MultiTabQA: Generating Tabular Answers for Multi-Table Question Answering
- Using counterfactual contrast to improve compositional generalization for multi-step quantitative reasoning
- The Dangers of trusting Stochastic Parrots: Faithfulness and Trust in Open-domain Conversational Question Answering
- Hybrid Hierarchical Retrieval for Open-Domain Question Answering
- Exploiting Abstract Meaning Representation for Open-Domain Question Answering
- IDOL: Indicator-oriented Logic Pre-training for Logical Reasoning
- A Survey for Efficient Open Domain Question Answering
- Generating Deep Questions with Commonsense Reasoning Ability from the Text by Disentangled Adversarial Inference
Resources and Evaluation (Virtual Poster)
Room: Pier 7&8
- PropSegmEnt: A Large-Scale Corpus for Proposition-Level Segmentation and Entailment Recognition
- LiveChat: A Large-Scale Personalized Dialogue Dataset Automatically Constructed from Live Streaming
- OpenPI-C: A Better Benchmark and Stronger Baseline for Open-Vocabulary State Tracking
- Table and Image Generation for Investigating Knowledge of Entities in Pre-trained Vision and Language Models
- We Understand Elliptical Sentences, and Language Models should Too: A New Dataset for Studying Ellipsis and its Interaction with Thematic Fit
- Measuring Consistency in Text-based Financial Forecasting Models
- Evaluate AMR Graph Similarity via Self-supervised Learning
- Echoes from Alexandria: A Large Resource for Multilingual Book Summarization
- Varta: A Large-Scale Headline-Generation Dataset for Indic Languages
- InfoSync: Information Synchronization across Multilingual Semi-structured Tables
- ISLTranslate: Dataset for Translating Indian Sign Language
- CrossSum: Beyond English-Centric Cross-Lingual Summarization for 1,500+ Language Pairs
- Open-WikiTable : Dataset for Open Domain Question Answering with Complex Reasoning over Table
- I run as fast as a rabbit, can you? A Multilingual Simile Dialogues Datasets
Semantics: Lexical (Virtual Poster)
Room: Pier 7&8
- ParaLS: Lexical Substitution via Pretrained Paraphraser
- Unsupervised Paraphrasing of Multiword Expressions
- Solving Cosine Similarity Underestimation between High Frequency Words by $\ell_2$ Norm Discounting
- Ambiguity Meets Uncertainty: Investigating Uncertainty Estimation for Word Sense Disambiguation
Semantics: Sentence-level Semantics, Textual Inference, and Other Areas (Virtual Poster)
Room: Pier 7&8
- Multi-Level Knowledge Distillation for Out-of-Distribution Detection in Text
- On Isotropy, Contextualization and Learning Dynamics of Contrastive-based Sentence Representation Learning
- Exploring Non-Verbal Predicates in Semantic Role Labeling: Challenges and Opportunities
- Incorporating Graph Information in Transformer-based AMR Parsing
- Entailment as Robust Self-Learner
- Composition-contrastive Learning for Sentence Embeddings
- The Best of Both Worlds: Combining Human and Machine Translations for Multilingual Semantic Parsing with Active Learning
- Cross-lingual AMR Aligner: Paying Attention to Cross-Attention
- Alleviating Over-smoothing for Unsupervised Sentence Representation
- Enhancing Language Representation with Constructional Information for Natural Language Understanding
- SETI: Systematicity Evaluation of Textual Inference
- Predicting Numerals in Text Using Nearest Neighbor Language Models
- Investigating Transformer-Guided Chaining for Interpretable Natural Logic Reasoning
Sentiment Analysis, Stylistic Analysis, and Argument Mining (Virtual Poster)
Room: Pier 7&8
- Sentiment Knowledge Enhanced Self-supervised Learning for Multimodal Sentiment Analysis
- Multilingual Multi-Figurative Language Detection
- Estimating the Uncertainty in Emotion Attributes using Deep Evidential Regression
- Bidirectional Generative Framework for Cross-domain Aspect-based Sentiment Analysis
Speech and Multimodality (demo) (Virtual Poster)
Room: Pier 7&8
Speech and Multimodality (Virtual Poster)
Room: Pier 7&8
- FastDiff 2: Revisiting and Incorporating GANs and Diffusion Models in High-Fidelity Speech Synthesis
- Werewolf Among Us: Multimodal Resources for Modeling Persuasion Behaviors in Social Deduction Games
- Multi-modal Action Chain Abductive Reasoning
- Visually-Enhanced Phrase Understanding
- Simple and Effective Unsupervised Speech Translation
- TableVLM: Multi-modal Pre-training for Table Structure Recognition
- Exploring Speaker-Related Information in Spoken Language Understanding for Better Speaker Diarization
- Deeply Coupled Cross-Modal Prompt Learning
Student Research Workshop (Virtual Poster)
Room: Pier 7&8
- [SRW] ChatGPT vs Human-authored Text: Insights into Controllable Text Summarization and Sentence Style Transfer
- [SRW] Prompt-based Zero-shot Text Classification with Conceptual Knowledge
- [SRW] Improving Portfolio Management with Signals from Financial News
- [SRW] Distractor Generation for Fill-in-the-Blank Exercises by Question Type
- [SRW] "When Words Fail, Emojis Prevail": A Novel Architecture for Generating Sarcastic Sentences With Emoji Using Valence Reversal and Semantic Incongruity
Summarization (Virtual Poster)
Room: Pier 7&8
- Unsupervised Summarization Re-ranking
- Interpretable Automatic Fine-grained Inconsistency Detection in Text Summarization
- Language Models for German Text Simplification: Overcoming Parallel Data Scarcity through Style-specific Pre-training
- NonFactS: NonFactual Summary Generation for Factuality Evaluation in Document Summarization
- Reference Matters: Benchmarking Factual Error Correction for Dialogue Summarization with Fine-grained Evaluation Framework
Syntax: Tagging, Chunking, and Parsing (Virtual Poster)
Room: Pier 7&8
- Unsupervised Mapping of Arguments of Deverbal Nouns to Their Corresponding Verbal Labels
- Probabilistic Transformer: A Probabilistic Dependency Model for Contextual Word Representation
- Convergence and Diversity in the Control Hierarchy
- Improving Grammar-based Sequence-to-Sequence Modeling with Decomposition and Constraints
- Another Dead End for Morphological Tags? Perturbed Inputs and Parsing
Theme: Reality Check (Virtual Poster)
Room: Pier 7&8
- Large Language Models Meet NL2Code: A Survey
- An Exploratory Study on Model Compression for Text-to-SQL
- Follow the leader(board) with confidence: Estimating p-values from a single test set with item and response variance
- Can Language Models Be Specific? How?
- Are Layout-Infused Language Models Robust to Layout Distribution Shifts? A Case Study with Scientific Documents
- A Comparative Analysis of the Effectiveness of Rare Tokens on Creative Expression using ramBERT
- A Survey on Zero Pronoun Translation
- A Systematic Study and Comprehensive Evaluation of ChatGPT on Benchmark Datasets
- Pulling Out All The Full Stops: Punctuation Sensitivity in Neural Machine Translation and Evaluation
Session 2
⏰ Time: July 10, 14:00-15:30 (EDT) Conference
⏰ Time: (18:00-19:30 UTC)
UTC /
Browser
Oral Presentations
Language Grounding to Vision, Robotics, and Beyond (Oral)
Room: Pier 4&5
- Cross2StrA: Unpaired Cross-lingual Image Captioning with Cross-lingual Cross-modal Structure-pivoted Alignment
- Why Did the Chicken Cross the Road? Rephrasing and Analyzing Ambiguous Questions in VQA
- VLN-Trans: Translator for the Vision and Language Navigation Agent
- Visually-augmented pretrained language models for NLP tasks without images
- Gloss-Free End-to-End Sign Language Translation
- VisText: A Benchmark for Semantically Rich Chart Captioning
Machine Learning for NLP (Oral)
Room: Metropolitan Centre
- Bridging the Gap between Decision and Logits in Decision-based Knowledge Distillation for Pre-trained Language Models
- f-Divergence Minimization for Sequence-Level Knowledge Distillation
- Patton: Language Model Pretraining on Text-Rich Networks
- Binary and Ternary Natural Language Generation
- Pruning Pre-trained Language Models Without Fine-Tuning
- Small Pre-trained Language Models Can be Fine-tuned as Large Models via Over-Parameterization
Machine Translation (Oral)
Room: Metropolitan West
- Extrinsic Evaluation of Machine Translation Metrics
- [TACL] FRMT: A Benchmark for Few-Shot Region-Aware Machine Translation
- Knowledge Transfer in Incremental Learning for Multilingual Neural Machine Translation
- Discourse-Centric Evaluation of Document-level Machine Translation with a New Densely Annotated Parallel Corpus of Novels
- BLEURT Has Universal Translations: An Analysis of Automatic Metrics by Minimum Risk Training
- xSIM++: An Improved Proxy to Bitext Mining Performance for Low-Resource Languages
Sentiment Analysis, Stylistic Analysis, and Argument Mining (Oral)
Room: Pier 2&3
- Guiding Computational Stance Detection with Expanded Stance Triangle Framework
- A New Direction in Stance Detection: Target-Stance Extraction in the Wild
- Node Placement in Argument Maps: Modeling Unidirectional Relations in High & Low-Resource Scenarios
- Topic-Guided Sampling For Data-Efficient Multi-Domain Stance Detection
- C-STANCE: A Large Dataset for Chinese Zero-Shot Stance Detection
- [CL] Comparing Selective Masking Methods for Depression Detection in Social Media
Syntax: Tagging, Chunking, and Parsing (Oral)
Room: Pier 7&8
- Holographic CCG Parsing
- Don't Parse, Choose Spans! Continuous and Discontinuous Constituency Parsing via Autoregressive Span Selection
- Contextual Distortion Reveals Constituency: Masked Language Models are Implicit Parsers
- Unsupervised Discontinuous Constituency Parsing with Mildly Context-Sensitive Grammars
- [CL] Cross-Lingual Transfer with Language-Specific Subnetworks for Low-Resource Dependency Parsing
- Hexatagging: Projective Dependency Parsing as Tagging
Theme: Reality Check (Oral)
Room: Metropolitan East
- Credible without Credit: Domain Experts Assess Generative Language Models
- What's the Meaning of Superhuman Performance in Today's NLU?
- Why Aren't We NER Yet? Artifacts of ASR Errors in Named Entity Recognition in Spontaneous Speech Transcripts
- Mind the Gap between the Application Track and the Real World
- Weaker Than You Think: A Critical Look at Weakly Supervised Learning
- On "Scientific Debt" in NLP: A Case for More Rigour in Language Model Pre-Training Research
Poster Presentations
Information Retrieval and Text Mining (Poster)
Room: Frontenac Ballroom and Queen's Quay
Machine Learning for NLP (Poster)
Room: Frontenac Ballroom and Queen's Quay
Spotlight Session
⏰ Time: July 10, 19:00-21:00 (EDT) Conference
⏰ Time: (23:00-01:00 UTC)
UTC /
Browser
Spotlight Presentations
Spotlight - Metropolitan Centre (Spotlight)
Room: Metropolitan Centre
- Track: Machine Learning for NLP
- On the Expressivity Role of LayerNorm in Transformers' Attention
- EmbedTextNet: Dimension Reduction with Weighted Reconstruction and Correlation Losses for Efficient Text Embedding
- A Memory Model for Question Answering from Streaming Data Supported by Rehearsal and Anticipation of Coreference Information
- CFL: Causally Fair Language Models Through Token-level Attribute Controlled Generation
- Text Augmentation Using Dataset Reconstruction for Low-Resource Classification
- Low-Rank Updates of pre-trained Weights for Multi-Task Learning
- AutoMoE: Heterogeneous Mixture-of-Experts with Adaptive Computation for Efficient Neural Machine Translation
- Neural Architecture Search for Parameter-Efficient Fine-tuning of Large Pre-trained Language Models
- Which Examples Should be Multiply Annotated? Active Learning When Annotators May Disagree
- B2T Connection: Serving Stability and Performance in Deep Transformers
- Reinforced Active Learning for Low-Resource, Domain-Specific, Multi-Label Text Classification
- PreQuant: A Task-agnostic Quantization Approach for Pre-trained Language Models
- Know Where You're Going: Meta-Learning for Parameter-Efficient Fine-Tuning
- Distilling Step-by-Step! Outperforming Larger Language Models with Less Training Data and Smaller Model Sizes
- LABO: Towards Learning Optimal Label Regularization via Bi-level Optimization
- Not Enough Data to Pre-train Your Language Model? MT to the Rescue!
- Exclusive Supermask Subnetwork Training for Continual Learning
- History repeats: Overcoming catastrophic forgetting for event-centric temporal knowledge graph completion
- Label Agnostic Pre-training for Zero-shot Text Classification
- ECOLA: Enhancing Temporal Knowledge Embeddings with Contextualized Language Representations
- Track: Large Language Models
- Recyclable Tuning for Continual Pre-training
- The Larger they are, the Harder they Fail: Language Models do not Recognize Identifier Swaps in Python
- Evaluating the Factual Consistency of Large Language Models Through News Summarization
- HELP ME THINK: A Simple Prompting Strategy for Non-experts to Create Customized Content with Models
- The Magic of IF: Investigating Causal Reasoning Abilities in Large Language Models of Code
- Membership Inference Attacks against Language Models via Neighbourhood Comparison
- Complementary Explanations for Effective In-Context Learning
- Nonparametric Masked Language Modeling
- Aligning Instruction Tasks Unlocks Large Language Models as Zero-Shot Relation Extractors
- Scaling Laws for BERT in Low-Resource Settings
- Large Language Models with Controllable Working Memory
- What In-Context Learning "Learns" In-Context: Disentangling Task Recognition and Task Learning
- Recipes for Sequential Pre-training of Multilingual Encoder and Seq2Seq Models
- Residual Prompt Tuning: improving prompt tuning with residual reparameterization
- Data-Efficient Finetuning Using Cross-Task Nearest Neighbors
- Honey, I Shrunk the Language: Language Model Behavior at Reduced Scale.
- Masked Latent Semantic Modeling: an Efficient Pre-training Alternative to Masked Language Modeling
- Beyond Positive Scaling: How Negation Impacts Scaling Trends of Language Models
- How does the task complexity of masked pretraining objectives affect downstream performance?
- Track: Generation
- Critic-Guided Decoding for Controlled Text Generation
- Revisiting the Architectures like Pointer Networks to Efficiently Improve the Next Word Distribution, Summarization Factuality, and Beyond
- Focus-aware Response Generation in Inquiry Conversation
- Nano: Nested Human-in-the-Loop Reward Learning for Few-shot Language Model Control
- Differentiable Instruction Optimization for Cross-Task Generalization
- Revisiting Sentence Union Generation as a Testbed for Text Consolidation
- PREADD: Prefix-Adaptive Decoding for Controlled Text Generation
- Efficient Out-of-Domain Detection for Sequence to Sequence Models
- Distilling Reasoning Capabilities into Smaller Language Models
- Language Modeling with Latent Situations
- Track: Summarization
- Unsupervised Summarization Re-ranking
- RISE: Leveraging Retrieval Techniques for Summarization Evaluation
- OpineSum: Entailment-based self-training for abstractive opinion summarization
- Towards Argument-Aware Abstractive Summarization of Long Legal Opinions with Summary Reranking
- Multi-Dimensional Evaluation of Text Summarization with In-Context Learning
- Improving Long Dialogue Summarization with Semantic Graph Representation
- Aspect-aware Unsupervised Extractive Opinion Summarization
- An Investigation of Evaluation Methods in Automatic Medical Note Generation
- Track: Machine Translation
- A Formal Perspective on Byte-Pair Encoding
- Lego-MT: Learning Detachable Models for Massively Multilingual Machine Translation
- What Knowledge Is Needed? Towards Explainable Memory for kNN-MT Domain Adaptation
- Towards Speech Dialogue Translation Mediating Speakers of Different Languages
- DUB: Discrete Unit Back-translation for Speech Translation
- Synthetic Pre-Training Tasks for Neural Machine Translation
- Towards Accurate Translation via Semantically Appropriate Application of Lexical Constraints
- Track: Information Extraction
- Dual-Gated Fusion with Prefix-Tuning for Multi-Modal Relation Extraction
- Multi-hop Evidence Retrieval for Cross-document Relation Extraction
- A Diffusion Model for Event Skeleton Generation
- Data Augmentation for Low-Resource Keyphrase Generation
- Silver Syntax Pre-training for Cross-Domain Relation Extraction
- Teamwork Is Not Always Good: An Empirical Study of Classifier Drift in Class-incremental Information Extraction
- Text Augmented Open Knowledge Graph Completion via Pre-Trained Language Models
- CoAug: Combining Augmentation of Labels and Labelling Rules
- Track: Language Grounding to Vision, Robotics, and Beyond
- Generate then Select: Open-ended Visual Question Answering Guided by World Knowledge
- Modularized Zero-shot VQA with Pre-trained Models
- Aerial Vision-and-Dialog Navigation
- Enhanced Chart Understanding via Visual Language Pre-training on Plot Table Pairs
- Towards Parameter-Efficient Integration of Pre-Trained Language Models In Temporal Video Grounding
- LMCap: Few-shot Multilingual Image Captioning by Retrieval Augmented Language Model Prompting
- Speaking the Language of Your Listener: Audience-Aware Adaptation via Plug-and-Play Theory of Mind
- Visual Coherence Loss for Coherent and Visually Grounded Story Generation
- I Spy a Metaphor: Large Language Models and Diffusion Models Co-Create Visual Metaphors
- Multimedia Generative Script Learning for Task Planning
- Track: Speech and Multimodality
- Listen, Decipher and Sign: Toward Unsupervised Speech-to-Sign Language Recognition
- Joint Speech Transcription and Translation: Pseudo-Labeling with Out-of-Distribution Data
- Visually-Enhanced Phrase Understanding
- FastDiff 2: Revisiting and Incorporating GANs and Diffusion Models in High-Fidelity Speech Synthesis
- Prosody-TTS: Improving Prosody with Masked Autoencoder and Conditional Diffusion Model For Expressive Text-to-Speech
Spotlight - Metropolitan East (Spotlight)
Room: Metropolitan East
- Track: Resources and Evaluation
- OpenPI-C: A Better Benchmark and Stronger Baseline for Open-Vocabulary State Tracking
- Exploiting Hierarchically Structured Categories in Fine-grained Chinese Named Entity Recognition
- Correction of Errors in Preference Ratings from Automated Metrics for Text Generation
- HeGeL: A Novel Dataset for Geo-Location from Hebrew Text
- An Annotated Dataset for Explainable Interpersonal Risk Factors of Mental Disturbance in Social Media Posts
- Echoes from Alexandria: A Large Resource for Multilingual Book Summarization
- A Unified Evaluation Framework for Novelty Detection and Accommodation in NLP with an Instantiation in Authorship Attribution
- RQUGE: Reference-Free Metric for Evaluating Question Generation by Answering the Question
- C-XNLI: Croatian Extension of XNLI Dataset
- ANALOGICAL - A Novel Benchmark for Long Text Analogy Evaluation in Large Language Models
- LEDA: a Large-Organization Email-Based Decision-Dialogue-Act Analysis Dataset
- Varta: A Large-Scale Headline-Generation Dataset for Indic Languages
- NusaCrowd: Open Source Initiative for Indonesian NLP Resources
- ORCA: A Challenging Benchmark for Arabic Language Understanding
- InfoSync: Information Synchronization across Multilingual Semi-structured Tables
- Take a Break in the Middle: Investigating Subgoals towards Hierarchical Script Generation
- ISLTranslate: Dataset for Translating Indian Sign Language
- PropSegmEnt: A Large-Scale Corpus for Proposition-Level Segmentation and Entailment Recognition
- Revisiting Sample Size Determination in Natural Language Understanding
- Track: Information Retrieval and Text Mining
- Dynamic Structured Neural Topic Model with Self-Attention Mechanism
- Discrete Prompt Optimization via Constrained Generation for Zero-shot Re-ranker
- DynaMiTE: Discovering Explosive Topic Evolutions with User Guidance
- Large Language Models are Built-in Autoregressive Search Engines
- Nonparametric Decoding for Generative Retrieval
- Recurrent Attention Networks for Long-text Modeling
- SamToNe: Improving Contrastive Loss for Dual Encoder Retrieval Models with Same Tower Negatives
- Task-aware Retrieval with Instructions
- Track: NLP Applications
- Detecting Adversarial Samples through Sharpness of Loss Landscape
- Score It All Together: A Multi-Task Learning Study on Automatic Scoring of Argumentative Essays
- An Exploration of Encoder-Decoder Approaches to Multi-Label Classification for Legal and Biomedical Text
- Prompt- and Trait Relation-aware Cross-prompt Essay Trait Scoring
- Towards Diverse and Effective Question-Answer Pair Generation from Children Storybooks
- Similarity-Based Content Scoring - A more Classroom-Suitable Alternative to Instance-Based Scoring?
- Sequential Path Signature Networks for Personalised Longitudinal Language Modeling
- Contrastive Training Improves Zero-Shot Classification of Semi-structured Documents
- GVdoc - Graph-based Visual DOcument Classification
- Distractor Generation based on Text2Text Language Models with Pseudo Kullback-Leibler Divergence Regulation
- Financial Numeric Extreme Labelling: A dataset and benchmarking
- Scientific Fact-Checking: A Survey of Resources and Approaches
- Understanding Programs by Exploiting (Fuzzing) Test Cases
- Track: Dialogue and Interactive Systems
- Dialogue Planning via Brownian Bridge Stochastic Process for Goal-directed Proactive Dialogue
- Injecting Comparison Skills in Task-Oriented Dialogue Systems for Database Search Results Disambiguation
- CausalDialogue: Modeling Utterance-level Causality in Conversations
- Multi-Domain Dialogue State Tracking with Disentangled Domain-Slot Attention
- Intent Discovery with Frame-guided Semantic Regularization and Augmentation
- DEnsity: Open-domain Dialogue Evaluation Metric using Density Estimation
- Multimodal Recommendation Dialog with Subjective Preference: A New Challenge and Benchmark
- Multi3NLU++: A Multilingual, Multi-Intent, Multi-Domain Dataset for Natural Language Understanding in Task-Oriented Dialogue
- Prompted LLMs as Chatbot Modules for Long Open-domain Conversation
- Zero-Shot Prompting for Implicit Intent Prediction and Recommendation with Commonsense Reasoning
- Boosting Distress Support Dialogue Responses with Motivational Interviewing Strategy
- Diverse Retrieval-Augmented In-Context Learning for Dialogue State Tracking
- The Whole Truth and Nothing But the Truth: Faithful and Controllable Dialogue Response Generation with Dataflow Transduction and Constrained Decoding
- Imagination is All You Need! Curved Contrastive Learning for Abstract Sequence Modeling Utilized on Long Short-Term Dialogue Planning
- Leveraging Explicit Procedural Instructions for Data-Efficient Action Prediction
- Track: Question Answering
- Improving Cross-task Generalization of Unified Table-to-text Models with Compositional Task Configurations
- Optimizing Test-Time Query Representations for Dense Retrieval
- TimelineQA: A Benchmark for Question Answering over Timelines
- Phrase Retrieval for Open Domain Conversational Question Answering with Conversational Dependency Modeling via Contrastive Learning
- DePlot: One-shot visual language reasoning by plot-to-table translation
- Coupling Large Language Models with Logic Programming for Robust and General Reasoning from Text
- World Models for Math Story Problems
- The Dangers of trusting Stochastic Parrots: Faithfulness and Trust in Open-domain Conversational Question Answering
- Expand, Rerank, and Retrieve: Query Reranking for Open-Domain Question Answering
- Hybrid Hierarchical Retrieval for Open-Domain Question Answering
- RobustQA: Benchmarking the Robustness of Domain Adaptation for Open-Domain Question Answering
- KoRC: Knowledge Oriented Reading Comprehension Benchmark for Deep Text Understanding
- Distinguish Before Answer: Generating Contrastive Explanation as Knowledge for Commonsense Question Answering
- An Empirical Comparison of LM-based Question and Answer Generation Methods
- Track: Sentiment Analysis, Stylistic Analysis, and Argument Mining
- Don't Lose Yourself! Empathetic Response Generation via Explicit Self-Other Awareness
- A Unified One-Step Solution for Aspect Sentiment Quad Prediction
- Few-shot Joint Multimodal Aspect-Sentiment Analysis Based on Generative Multimodal Prompt
- A Simple Yet Strong Domain-Agnostic De-bias Method for Zero-Shot Sentiment Classification
- TransESC: Smoothing Emotional Support Conversation via Turn-Level State Transition
- Multilingual Multi-Figurative Language Detection
- Track: Computational Social Science and Cultural Analytics
- Counterfactual Probing for the Influence of Affect and Specificity on Intergroup Bias
- Measuring Intersectional Biases in Historical Documents
- It's not Sexually Suggestive; It's Educative | Separating Sex Education from Suggestive Content on TikTok videos
- Contrastive Learning of Sociopragmatic Meaning in Social Media
- Words as Gatekeepers: Measuring Discipline-specific Terms and Meanings in Scholarly Publications
- Modeling Cross-Cultural Pragmatic Inference with Codenames Duet
- Causal Matching with Text Embeddings: A Case Study in Estimating the Causal Effects of Peer Review Policies
- Dramatic Conversation Disentanglement
- Responsibility Perspective Transfer for Italian Femicide News
- On Text-based Personality Computing: Challenges and Future Directions
Spotlight - Metropolitan West (Spotlight)
Room: Metropolitan West
- Track: Phonology, Morphology, and Word Segmentation
- Subword Segmental Machine Translation: Unifying Segmentation and Target Sentence Generation
- An Investigation of Noise in Morphological Inflection
- Track: Syntax: Tagging, Chunking, and Parsing
- Pre-Trained Language-Meaning Models for Multilingual Parsing and Generation
- Unsupervised Mapping of Arguments of Deverbal Nouns to Their Corresponding Verbal Labels
- Track: Semantics: Lexical
- DMLM: Descriptive Masked Language Modeling
- Ambiguity Meets Uncertainty: Investigating Uncertainty Estimation for Word Sense Disambiguation
- A Self-Supervised Integration Method of Pretrained Language Models and Word Definitions
- Unsupervised Paraphrasing of Multiword Expressions
- Together We Make Sense--Learning Meta-Sense Embeddings
- Solving Cosine Similarity Underestimation between High Frequency Words by $\ell_2$ Norm Discounting
- Improving Diachronic Word Sense Induction with a Nonparametric Bayesian method
- Track: Semantics: Sentence-level Semantics, Textual Inference, and Other Areas
- On Isotropy, Contextualization and Learning Dynamics of Contrastive-based Sentence Representation Learning
- Exploring Non-Verbal Predicates in Semantic Role Labeling: Challenges and Opportunities
- Incorporating Graph Information in Transformer-based AMR Parsing
- Align-then-Enhance: Multilingual Entailment Graph Enhancement with Soft Predicate Alignment
- Track: Linguistic Theories, Cognitive Modeling, and Psycholinguistics
- Unsupervised Semantic Variation Prediction using the Distribution of Sibling Embeddings
- Categorial grammar induction from raw data
- Language acquisition: do children and language models follow similar learning stages?
- Automatic Readability Assessment for Closely Related Languages
- Track: Discourse and Pragmatics
- $2*n$ is better than $n^2$: Decomposing Event Coreference Resolution into Two Tractable Problems
- $2*n$ is better than $n^2$: Decomposing Event Coreference Resolution into Two Tractable Problems
- How Well Do Large Language Models Perform on Faux Pas Tests?
- Distinguishing Address vs. Reference Mentions of Personal Names in Text
- PragmatiCQA: A Dataset for Pragmatic Question Answering in Conversations
- A Match Made in Heaven: A Multi-task Framework for Hyperbole and Metaphor Detection
- Towards Generative Event Factuality Prediction
- Discourse Analysis via Questions and Answers: Parsing Dependency Structures of Questions Under Discussion
- Track: Linguistic Diversity
- AxomiyaBERTa: A Phonologically-aware Transformer Model for Assamese
- SERENGETI: Massively Multilingual Language Models for Africa
- Verifying Annotation Agreement without Multiple Experts: A Case Study with Gujarati SNACS
- Track: Multilingualism and Cross-Lingual NLP
- X-RiSAWOZ: High-Quality End-to-End Multilingual Dialogue Datasets and Few-shot Agents
- Can Cross-Lingual Transferability of Multilingual Transformers Be Activated Without End-Task Data?
- Language Agnostic Multilingual Information Retrieval with Contrastive Learning
- Why Does Zero-Shot Cross-Lingual Generation Fail? An Explanation and a Solution
- Cross-Lingual Retrieval Augmented Prompt for Low-Resource Languages
- Predicting Human Translation Difficulty Using Automatic Word Alignment
- Automatic Identification of Code-Switching Functions in Speech Transcripts
- Tokenization Impacts Multilingual Language Modeling: Assessing Vocabulary Allocation and Overlap Across Languages
- Code-Switched Text Synthesis in Unseen Language Pairs
- Exploring Anisotropy and Outliers in Multilingual Language Models for Cross-Lingual Semantic Sentence Similarity
- Frustratingly Easy Label Projection for Cross-lingual Transfer
- TADA : Task Agnostic Dialect Adapters for English
- Mini-Model Adaptation: Efficiently Extending Pretrained Models to New Languages via Aligned Shallow Training
- Track: Interpretability and Analysis of Models for NLP
- Explanation Regeneration via Information Bottleneck
- Characterizing the Impacts of Instances on Robustness
- Counterfactuals of Counterfactuals: a back-translation-inspired approach to analyse counterfactual editors
- Conformal Nucleus Sampling
- Fighting Bias With Bias: Promoting Model Robustness by Amplifying Dataset Biases
- Robustness of Learning from Task Instructions
- COCKATIEL: COntinuous Concept ranKed ATtribution with Interpretable ELements for explaining neural net classifiers on NLP
- Figurative Language Processing: A Linguistically Informed Feature Analysis of the Behavior of Language Models and Humans
- Model Interpretability and Rationale Extraction by Input Mask Optimization
- Layerwise universal adversarial attack on NLP models
- Inducing Character-level Structure in Subword-based Language Models with Type-level Interchange Intervention Training
- SenteCon: Leveraging Lexicons to Learn Human-Interpretable Language Representations
- HyHTM: Hyperbolic Geometry-based Hierarchical Topic Model
- Adversarial Robustness of Prompt-based Few-Shot Learning for Natural Language Understanding
- Robust Natural Language Understanding with Residual Attention Debiasing
- Transformer Language Models Handle Word Frequency in Prediction Head
- Is Continuous Prompt a Combination of Discrete Prompts? Towards a Novel View for Interpreting Continuous Prompts
- Claim-Dissector: An Interpretable Fact-Checking System with Joint Re-ranking and Veracity Prediction
- PromptAttack: Probing Dialogue State Trackers with Adversarial Prompts
- Track: Ethics and NLP
- Disagreement Matters: Preserving Label Diversity by Jointly Modeling Item and Annotator Label Distributions with DisCo
- Bias Beyond English: Counterfactual Tests for Bias in Sentiment Analysis in Four Languages
- Debiasing should be Good and Bad: Measuring the Consistency of Debiasing Techniques in Language Models
- FORK: A Bite-Sized Test Set for Probing Culinary Cultural Biases in Commonsense Reasoning Models
- Foveate, Attribute, and Rationalize: Towards Physically Safe and Trustworthy AI
- Shielded Representations: Protecting Sensitive Attributes Through Iterative Gradient-Based Projection
- T2IAT: Measuring Valence and Stereotypical Biases in Text-to-Image Generation
- Track: Theme: Reality Check
- An Exploratory Study on Model Compression for Text-to-SQL
- It Takes Two to Tango: Navigating Conceptualizations of NLP Tasks and Measurements of Performance
- Reimagining Retrieval Augmented Language Models for Answering Queries
- Follow the leader(board) with confidence: Estimating p-values from a single test set with item and response variance
- Can Language Models Be Specific? How?
- Are Layout-Infused Language Models Robust to Layout Distribution Shifts? A Case Study with Scientific Documents
- A Comparative Analysis of the Effectiveness of Rare Tokens on Creative Expression using ramBERT
- The Devil is in the Details: On the Pitfalls of Event Extraction Evaluation
- A Call for Standardization and Validation of Text Style Transfer Evaluation
- This prompt is measuring <mask>: evaluating bias evaluation in language models
- A Systematic Study and Comprehensive Evaluation of ChatGPT on Benchmark Datasets
- Few-shot Fine-tuning vs. In-context Learning: A Fair Comparison and Evaluation
- Reproducibility in NLP: What Have We Learned from the Checklist?
- GUMSum: Multi-Genre Data and Evaluation for English Abstractive Summarization
- Towards Reasoning in Large Language Models: A Survey
- Towards Reasoning in Large Language Models: A Survey
Timezone: Conference (Toronto) UTC Browser
Demo Session 3
⏰ Time: July 11, 09:00-10:30 (EDT) Conference
⏰ Time: (13:00-14:30 UTC)
UTC /
Browser
Poster Presentations
Dialogue and Interactive Systems (demo) (Poster)
Room: Frontenac Ballroom and Queen's Quay
Information Extraction (demo) (Poster)
Room: Frontenac Ballroom and Queen's Quay
Linguistic Diversity (demo) (Poster)
Room: Frontenac Ballroom and Queen's Quay
Machine Translation (demo) (Poster)
Room: Frontenac Ballroom and Queen's Quay
NLP Applications (demo) (Poster)
Room: Frontenac Ballroom and Queen's Quay
- [Demo] CARE: Collaborative AI-Assisted Reading Environment
- [Demo] LaTeX2Solver: a Hierarchical Semantic Parsing of LaTeX Document into Code for an Assistive Optimization Modeling Application
- [Demo] Disease Network Constructor: a Pathway Extraction and Visualization
- [Demo] Effidit: An Assistant for Improving Writing Efficiency
Resources and Evaluation (demo) (Poster)
Room: Frontenac Ballroom and Queen's Quay
Speech and Multimodality (demo) (Poster)
Room: Frontenac Ballroom and Queen's Quay
Demo Session 4
⏰ Time: July 11, 11:00-12:30 (EDT) Conference
⏰ Time: (15:00-16:30 UTC)
UTC /
Browser
Poster Presentations
Information Extraction (demo) (Poster)
Room: Frontenac Ballroom and Queen's Quay
Machine Learning for NLP (demo) (Poster)
Room: Frontenac Ballroom and Queen's Quay
Machine Translation (demo) (Poster)
Room: Frontenac Ballroom and Queen's Quay
Multilingualism and Cross-Lingual NLP (demo) (Poster)
Room: Frontenac Ballroom and Queen's Quay
Demo Session 5
⏰ Time: July 11, 16:15-17:45 (EDT) Conference
⏰ Time: (20:15-21:45 UTC)
UTC /
Browser
Poster Presentations
Interpretability and Analysis of Models for NLP (demo) (Poster)
Room: Frontenac Ballroom and Queen's Quay
Large Language Models (demo) (Poster)
Room: Frontenac Ballroom and Queen's Quay
Question Answering (demo) (Poster)
Room: Frontenac Ballroom and Queen's Quay
Resources and Evaluation (demo) (Poster)
Room: Frontenac Ballroom and Queen's Quay
Poster Session 3
⏰ Time: July 11, 09:00-10:30 (EDT) Conference
⏰ Time: (13:00-14:30 UTC)
UTC /
Browser
Poster Presentations
Dialogue and Interactive Systems (Poster)
Room: Frontenac Ballroom and Queen's Quay
- MPCHAT: Towards Multimodal Persona-Grounded Conversation
- Span-Selective Linear Attention Transformers for Effective and Robust Schema-Guided Dialogue State Tracking
- Query-Efficient Black-Box Red Teaming via Bayesian Optimization
- SIMMC-VR: A Task-oriented Multimodal Dialog Dataset with Situated and Immersive VR Streams
Discourse and Pragmatics (Poster)
Room: Frontenac Ballroom and Queen's Quay
Ethics and NLP (Poster)
Room: Frontenac Ballroom and Queen's Quay
Generation (Poster)
Room: Frontenac Ballroom and Queen's Quay
- Enhancing Dialogue Generation via Dynamic Graph Knowledge Aggregation
- Attractive Storyteller: Stylized Visual Storytelling with Unpaired Text
- An Extensible Plug-and-Play Method for Multi-Aspect Controllable Text Generation
- With a Little Push, NLI Models can Robustly and Efficiently Predict Faithfulness
- LENS: A Learnable Evaluation Metric for Text Simplification
- BOLT: Fast Energy-based Controlled Text Generation with Tunable Biases
- DuNST: Dual Noisy Self Training for Semi-Supervised Controllable Text Generation
- NEUROSTRUCTURAL DECODING: Neural Text Generation with Structural Constraints
Information Extraction (Poster)
Room: Frontenac Ballroom and Queen's Quay
- Joint End-to-end Semantic Proto-role Labeling
- Jointprop: Joint Semi-supervised Learning for Entity and Relation Extraction with Heterogeneous Graph-based Propagation
- Learning "O" Helps for Learning More: Handling the Unlabeled Entity Problem for Class-incremental NER
- WebIE: Faithful and Robust Information Extraction on the Web
- FormNetV2: Multimodal Graph Contrastive Learning for Form Document Information Extraction
- Automatic Creation of Named Entity Recognition Datasets by Querying Phrase Representations
- S2ynRE: Two-stage Self-training with Synthetic data for Low-resource Relation Extraction
Information Retrieval and Text Mining (Poster)
Room: Frontenac Ballroom and Queen's Quay
Interpretability and Analysis of Models for NLP (Poster)
Room: Frontenac Ballroom and Queen's Quay
- Towards Robust Low-Resource Fine-Tuning with Multi-View Compressed Representations
- Randomized Smoothing with Masked Inference for Adversarially Robust Text Classifications
- Exploring Lottery Prompts for Pre-trained Language Models
- How to Plant Trees in Language Models: Data and Architectural Effects on the Emergence of Syntactic Inductive Biases
- Being Right for Whose Right Reasons?
- infoVerse: A Universal Framework for Dataset Characterization with Multidimensional Meta-information
Language Grounding to Vision, Robotics, and Beyond (Poster)
Room: Frontenac Ballroom and Queen's Quay
Large Language Models (Poster)
Room: Frontenac Ballroom and Queen's Quay
Machine Learning for NLP (Poster)
Room: Frontenac Ballroom and Queen's Quay
- Prefix Propagation: Parameter-Efficient Tuning for Long Sequences
- Multi-CLS BERT: An Efficient Alternative to Traditional Ensembling
- PESCO: Prompt-enhanced Self Contrastive Learning for Zero-shot Text Classification
- Parameter-Efficient Fine-Tuning without Introducing New Latency
- Counterfactual Active Learning for Out-of-Distribution Generalization
- Cold-Start Data Selection for Better Few-shot Language Model Fine-tuning: A Prompt-based Uncertainty Propagation Approach
- PeerDA: Data Augmentation via Modeling Peer Relation for Span Identification Tasks
Machine Translation (Poster)
Room: Frontenac Ballroom and Queen's Quay
Multilingualism and Cross-Lingual NLP (Poster)
Room: Frontenac Ballroom and Queen's Quay
NLP Applications (Poster)
Room: Frontenac Ballroom and Queen's Quay
- Self-Edit: Fault-Aware Code Editor for Code Generation
- A Simple and Effective Framework for Strict Zero-Shot Hierarchical Classification
- A Causal Framework to Quantify the Robustness of Mathematical Reasoning with Language Models
- MIReAD: Simple Method for Learning High-quality Representations from Scientific Documents
Phonology, Morphology, and Word Segmentation (Poster)
Room: Frontenac Ballroom and Queen's Quay
Question Answering (Poster)
Room: Frontenac Ballroom and Queen's Quay
- Covering Uncommon Ground: Gap-Focused Question Generation for Answer Assessment
- Do I have the Knowledge to Answer? Investigating Answerability of Knowledge Base Questions
- Query Structure Modeling for Inductive Logical Reasoning Over Knowledge Graphs
- Interleaving Retrieval with Chain-of-Thought Reasoning for Knowledge-Intensive Multi-Step Questions
- Single Sequence Prediction over Reasoning Graphs for Multi-hop QA
- Few-shot Reranking for Multi-hop QA via Language Model Prompting
Resources and Evaluation (Poster)
Room: Frontenac Ballroom and Queen's Quay
- An Open Dataset and Model for Language Identification
- Rethinking Annotation: Can Language Learners Contribute?
- Environmental Claim Detection
- Towards standardizing Korean Grammatical Error Correction: Datasets and Annotation
- TeCS: A Dataset and Benchmark for Tense Consistency of Machine Translation
- FERMAT: An Alternative to Accuracy for Numerical Reasoning
- Towards Leaving No Indic Language Behind: Building Monolingual Corpora, Benchmark and Models for Indic Languages
- ELQA: A Corpus of Metalinguistic Questions and Answers about English
- Transfer and Active Learning for Dissonance Detection: Addressing the Rare-Class Challenge
- VSTAR: A Video-grounded Dialogue Dataset for Situated Semantic Understanding with Scene and Topic Transitions
- DarkBERT: A Language Model for the Dark Side of the Internet
- READIN: A Chinese Multi-Task Benchmark with Realistic and Diverse Input Noises
Semantics: Lexical (Poster)
Room: Frontenac Ballroom and Queen's Quay
Semantics: Sentence-level Semantics, Textual Inference, and Other Areas (Poster)
Room: Frontenac Ballroom and Queen's Quay
Speech and Multimodality (Poster)
Room: Frontenac Ballroom and Queen's Quay
Syntax: Tagging, Chunking, and Parsing (Poster)
Room: Frontenac Ballroom and Queen's Quay
Poster Session 4
⏰ Time: July 11, 11:00-12:30 (EDT) Conference
⏰ Time: (15:00-16:30 UTC)
UTC /
Browser
Poster Presentations
Computational Social Science and Cultural Analytics (Poster)
Room: Frontenac Ballroom and Queen's Quay
Dialogue and Interactive Systems (Poster)
Room: Frontenac Ballroom and Queen's Quay
- Controllable Mixed-Initiative Dialogue Generation through Prompting
- ClarifyDelphi: Reinforced Clarification Questions with Defeasibility Rewards for Social and Moral Situations
- Privacy-Preserving Domain Adaptation of Semantic Parsers
- ChatGPT for Zero-shot Dialogue State Tracking: A Solution or an Opportunity?
- PeaCoK: Persona Commonsense Knowledge for Consistent and Engaging Narratives
- BREAK: Breaking the Dialogue State Tracking Barrier with Beam Search and Re-ranking
- Learning to Generate Equitable Text in Dialogue from Biased Training Data
Discourse and Pragmatics (Poster)
Room: Frontenac Ballroom and Queen's Quay
Ethics and NLP (Poster)
Room: Frontenac Ballroom and Queen's Quay
- On Second Thought, Let's Not Think Step by Step! Bias and Toxicity in Zero-Shot Reasoning
- Language Models Get a Gender Makeover: Mitigating Gender Bias with Few-Shot Data Interventions
- MISGENDERED: Limits of Large Language Models in Understanding Pronouns
- Improving Gender Fairness of Pre-Trained Language Models without Catastrophic Forgetting
Generation (Poster)
Room: Frontenac Ballroom and Queen's Quay
- A Systematic Study of Knowledge Distillation for Natural Language Generation with Pseudo-Target Training
- A Natural Bias for Language Generation Models
- Efficient Transformers with Dynamic Token Pooling
- Unsupervised Melody-to-Lyrics Generation
- Seen to Unseen: Exploring Compositional Generalization of Multi-Attribute Controllable Dialogue Generation
Information Extraction (Poster)
Room: Frontenac Ballroom and Queen's Quay
- Did the Models Understand Documents? Benchmarking Models for Language Understanding in Document-Level Relation Extraction
- Prompts Can Play Lottery Tickets Well: Achieving Lifelong Information Extraction via Lottery Prompt Tuning
- SPEECH: Structured Prediction with Energy-Based Event-Centric Hyperspheres
Information Retrieval and Text Mining (Poster)
Room: Frontenac Ballroom and Queen's Quay
Interpretability and Analysis of Models for NLP (Poster)
Room: Frontenac Ballroom and Queen's Quay
Language Grounding to Vision, Robotics, and Beyond (Poster)
Room: Frontenac Ballroom and Queen's Quay
Large Language Models (Poster)
Room: Frontenac Ballroom and Queen's Quay
- A Measure-Theoretic Characterization of Tight Language Models
- Parallel Context Windows for Large Language Models
- Two-Stage Fine-Tuning for Improved Bias and Variance for Large Pretrained Language Models
- Self-Instruct: Aligning Language Models with Self-Generated Instructions
- Knowledge Unlearning for Mitigating Privacy Risks in Language Models
- Data Curation Alone Can Stabilize In-context Learning
Linguistic Theories, Cognitive Modeling, and Psycholinguistics (Poster)
Room: Frontenac Ballroom and Queen's Quay
- What does the Failure to Reason with "Respectively'' in Zero/Few-Shot Settings Tell Us about Language Models?
- Just Like a Human Would, Direct Access to Sarcasm Augmented with Potential Result and Reaction
- How poor is the stimulus? Evaluating hierarchical generalization in neural networks trained on child-directed speech
Machine Learning for NLP (Poster)
Room: Frontenac Ballroom and Queen's Quay
- Class based Influence Functions for Error Detection
- Syntax and Geometry of Information
- FiD-ICL: A Fusion-in-Decoder Approach for Efficient In-Context Learning
- Distill or Annotate? Cost-Efficient Fine-Tuning of Compact Models
- Learning Symbolic Rules over Abstract Meaning Representations for Textual Reinforcement Learning
- LM-CPPF: Paraphrasing-Guided Data Augmentation for Contrastive Prompt-Based Few-Shot Fine-Tuning
Machine Translation (Poster)
Room: Frontenac Ballroom and Queen's Quay
Multilingualism and Cross-Lingual NLP (Poster)
Room: Frontenac Ballroom and Queen's Quay
NLP Applications (Poster)
Room: Frontenac Ballroom and Queen's Quay
- MUSTIE: Multimodal Structural Transformer for Web Information Extraction
- Faking Fake News for Real Fake News Detection: Propaganda-Loaded Training Data Generation
- When Does Aggregating Multiple Skills with Multi-Task Learning Work? A Case Study in Financial NLP
- Rule By Example: Harnessing Logical Rules for Explainable Hate Speech Detection
- Fact-Checking Complex Claims with Program-Guided Reasoning
- Unsupervised Subtitle Segmentation with Masked Language Models
- A Cognitive Stimulation Dialogue System with Multi-source Knowledge Fusion for Elders with Cognitive Impairment
- ACTC: Active Threshold Calibration for Cold-Start Knowledge Graph Completion
Question Answering (Poster)
Room: Frontenac Ballroom and Queen's Quay
Resources and Evaluation (Poster)
Room: Frontenac Ballroom and Queen's Quay
- Distilling Script Knowledge from Large Language Models for Constrained Language Planning
- HistRED: A Historical Document-Level Relation Extraction Dataset
- Facilitating Fine-grained Detection of Chinese Toxic Language: Hierarchical Taxonomy, Resources, and Benchmarks
- Hints on the data for language modeling of synthetic languages with transformers
- How to Distill your BERT: An Empirical Study on the Impact of Weight Initialisation and Distillation Objectives
- BUMP: A Benchmark of Unfaithful Minimal Pairs for Meta-Evaluation of Faithfulness Metrics
- FACTIFY-5WQA: 5W Aspect-based Fact Verification through Question Answering
- A Better Way to Do Masked Language Model Scoring
- Automatic Annotation of Direct Speech in Written French Narratives
- On the Evaluation of Neural Selective Prediction Methods for Natural Language Processing
- A Textual Dataset for Situated Proactive Response Selection
Semantics: Lexical (Poster)
Room: Frontenac Ballroom and Queen's Quay
Sentiment Analysis, Stylistic Analysis, and Argument Mining (Poster)
Room: Frontenac Ballroom and Queen's Quay
Speech and Multimodality (Poster)
Room: Frontenac Ballroom and Queen's Quay
Summarization (Poster)
Room: Frontenac Ballroom and Queen's Quay
Syntax: Tagging, Chunking, and Parsing (Poster)
Room: Frontenac Ballroom and Queen's Quay
Theme: Reality Check (Poster)
Room: Frontenac Ballroom and Queen's Quay
- Considerations for meaningful sign language machine translation based on glosses
- Dialect-robust Evaluation of Generated Text
- What about "em"? How Commercial Machine Translation Fails to Handle (Neo-)Pronouns
- Dealing with Semantic Underspecification in Multimodal NLP
- Forgotten Knowledge: Examining the Citational Amnesia in NLP
- Theory-Grounded Computational Text Analysis
- What Do NLP Researchers Believe? Results of the NLP Community Metasurvey
Poster Session 5
⏰ Time: July 11, 16:15-17:45 (EDT) Conference
⏰ Time: (20:15-21:45 UTC)
UTC /
Browser
Poster Presentations
Computational Social Science and Cultural Analytics (Poster)
Room: Frontenac Ballroom and Queen's Quay
Dialogue and Interactive Systems (Poster)
Room: Frontenac Ballroom and Queen's Quay
Generation (Poster)
Room: Frontenac Ballroom and Queen's Quay
Information Extraction (Poster)
Room: Frontenac Ballroom and Queen's Quay
- PromptNER: Prompt Locating and Typing for Named Entity Recognition
- ACLM: A Selective-Denoising based Generative Data Augmentation Approach for Low-Resource Complex NER
- The Role of Global and Local Context in Named Entity Recognition
- Event Extraction as Question Generation and Answering
- RED<sup>FM</sup>: a Filtered and Multilingual Relation Extraction Dataset
Interpretability and Analysis of Models for NLP (Poster)
Room: Frontenac Ballroom and Queen's Quay
Large Language Models (Poster)
Room: Frontenac Ballroom and Queen's Quay
Linguistic Theories, Cognitive Modeling, and Psycholinguistics (Poster)
Room: Frontenac Ballroom and Queen's Quay
Machine Learning for NLP (Poster)
Room: Frontenac Ballroom and Queen's Quay
- HiPool: Modeling Long Documents Using Graph Neural Networks
- DrBERT: A Robust Pre-trained Model in French for Biomedical and Clinical domains
- KALM: Knowledge-Aware Integration of Local, Document, and Global Contexts for Long Document Understanding
- Plug-and-Play Knowledge Injection for Pre-trained Language Models
Machine Translation (Poster)
Room: Frontenac Ballroom and Queen's Quay
NLP Applications (Poster)
Room: Frontenac Ballroom and Queen's Quay
- Target-Based Offensive Language Identification
- Multilingual Knowledge Graph Completion with Language-Sensitive Multi-Graph Attention
- HAHE: Hierarchical Attention for Hyper-Relational Knowledge Graphs in Global and Local Level
- Text-to-SQL Error Correction with Language Models of Code
- Characterization of Stigmatizing Language in Medical Records
- GreenKGC: A Lightweight Knowledge Graph Completion Method
- Neural Machine Translation for Mathematical Formulae
Question Answering (Poster)
Room: Frontenac Ballroom and Queen's Quay
Resources and Evaluation (Poster)
Room: Frontenac Ballroom and Queen's Quay
Semantics: Sentence-level Semantics, Textual Inference, and Other Areas (Poster)
Room: Frontenac Ballroom and Queen's Quay
Sentiment Analysis, Stylistic Analysis, and Argument Mining (Poster)
Room: Frontenac Ballroom and Queen's Quay
- Reasoning Implicit Sentiment with Chain-of-Thought Prompting
- Similarity-weighted Construction of Contextualized Commonsense Knowledge Graphs for Knowledge-intense Argumentation Tasks
- ArgAnalysis35K : A large-scale dataset for Argument Quality Analysis
- Trading Syntax Trees for Wordpieces: Target-oriented Opinion Words Extraction with Wordpieces and Aspect Enhancement
Summarization (Poster)
Room: Frontenac Ballroom and Queen's Quay
Theme: Reality Check (Poster)
Room: Frontenac Ballroom and Queen's Quay
Session 3
⏰ Time: July 11, 09:00-10:15 (EDT) Conference
⏰ Time: (13:00-14:15 UTC)
UTC /
Browser
Oral Presentations
Computational Social Science and Cultural Analytics (Oral)
Room: Pier 2&3
- Conflicts, Villains, Resolutions: Towards models of Narrative Media Framing
- My side, your side and the evidence: Discovering aligned actor groups and the narratives they weave
- Grounding Characters and Places in Narrative Text
- Your spouse needs professional help: Determining the Contextual Appropriateness of Messages through Modeling Social Relationships
- Understanding Client Reactions in Online Mental Health Counseling
Dialogue and Interactive Systems (Oral)
Room: Metropolitan West
- GIFT: Graph-Induced Fine-Tuning for Multi-Party Conversation Understanding
- Envisioning Future from the Past: Hierarchical Duality Learning for Multi-Turn Dialogue Generation
- Schema-Guided User Satisfaction Modeling for Task-Oriented Dialogues
- Divide, Conquer, and Combine: Mixture of Semantic-Independent Experts for Zero-Shot Dialogue State Tracking
- Towards Boosting the Open-Domain Chatbot with Human Feedback
- DiffusEmp: A Diffusion Model-Based Framework with Multi-Grained Control for Empathetic Response Generation
Industry (Oral)
Room: Pier 4&5
- [Industry] pNLP-Mixer: an Efficient all-MLP Architecture for Language
- [Industry] BADGE: Speeding Up BERT Inference after Deployment via Block-wise Bypasses and Divergence-based Early Exiting
- [Industry] K-pop and fake facts: from texts to smart alerting for maritime security
- [Industry] Context-Aware Query Rewriting for Improving Users' Search Experience on E-commerce Websites
- [Industry] Large Scale Generative Multimodal Attribute Extraction for E-commerce Attributes
- [Industry] Annotating Research Infrastructure in Scientific Papers: An NLP-driven Approach
Interpretability and Analysis of Models for NLP (Oral)
Room: Metropolitan East
- Incorporating Attribution Importance for Improving Faithfulness Metrics
- Generalizing Backpropagation for Gradient-Based Interpretability
- CREST: A Joint Framework for Rationalization and Counterfactual Text Generation
- SCOTT: Self-Consistent Chain-of-Thought Distillation
- Are Machine Rationales (Not) Useful to Humans? Measuring and Improving Human Utility of Free-text Rationales
- Faithfulness Tests for Natural Language Explanations
Large Language Models (Oral)
Room: Metropolitan Centre
- Glot500: Scaling Multilingual Corpora and Language Models to 500 Languages
- mCLIP: Multilingual CLIP via Cross-lingual Transfer
- Preserving Commonsense Knowledge from Pre-trained Language Models via Causal Inference
- Self-Adaptive In-Context Learning: An Information Compression Perspective for In-Context Example Selection and Ordering
- Pre-Training to Learn in Context
- ReAugKD: Retrieval-Augmented Knowledge Distillation For Pre-trained Language Models
Linguistic Diversity (Oral)
Room: Pier 7&8
- Script Normalization for Unconventional Writing of Under-Resourced Languages in Bilingual Communities
- Small Data, Big Impact: Leveraging Minimal Data for Effective Machine Translation
- Question-Answering in a Low-resourced Language: Benchmark Dataset and Models for Tigrinya
- MasakhaPOS: Part-of-Speech Tagging for Typologically Diverse African languages
- An (unhelpful) guide to selecting the best ASR architecture for your under-resourced language
- NollySenti: Leveraging Transfer Learning and Machine Translation for Nigerian Movie Sentiment Classification
Poster Presentations
Discourse and Pragmatics (Poster)
Room: Frontenac Ballroom and Queen's Quay
Phonology, Morphology, and Word Segmentation (Poster)
Room: Frontenac Ballroom and Queen's Quay
Student Research Workshop (Poster)
Room: Frontenac Ballroom and Queen's Quay
- [SRW] Constructing Multilingual Code Search Dataset Using Neural Machine Translation
- [SRW] Multimodal Neural Machine Translation Using Synthetic Images Transformed by Latent Diffusion Model
- [SRW] Predicting Human Translation Difficulty Using Automatic Word Alignment
- [SRW] Is Anisotropy Inherent to Transformers?
- [SRW] Geometric Locality of Entity Embeddings in Masked Language Models
- [SRW] Native Language Prediction from Gaze: a Reproducibility Study
- [SRW] Sudden Semantic Shifts in Swedish NATO discourse
- [SRW] Choosing What to Mask: More Informed Masking for Multimodal Machine Translation
- [SRW] Transformer Language Models Handle Word Frequency in Prediction Head
- [SRW] Jamp: Controlled Japanese Temporal Inference Dataset for Evaluating Generalization Capacity of Language Models
Syntax: Tagging, Chunking, and Parsing (Poster)
Room: Frontenac Ballroom and Queen's Quay
Session 4
⏰ Time: July 11, 11:00-11:45 (EDT) Conference
⏰ Time: (15:00-15:45 UTC)
UTC /
Browser
Oral Presentations
Language Grounding to Vision, Robotics, and Beyond (Oral)
Room: Pier 4&5
Large Language Models (Oral)
Room: Metropolitan Centre
- RetroMAE-2: Duplex Masked Auto-Encoder For Pre-Training Retrieval-Oriented Language Models
- Augmentation-Adapted Retriever Improves Generalization of Language Models as Generic Plug-In
- Pre-trained Language Models Can be Fully Zero-Shot Learners
- Multilingual LLMs are Better Cross-lingual In-context Learners with Alignment
- Surface-Based Retrieval Reduces Perplexity of Retrieval-Augmented Language Models
- Understanding In-Context Learning via Supportive Pretraining Data
Resources and Evaluation (Oral)
Room: Metropolitan East
- Evaluating Open-Domain Dialogues in Latent Space with Next Sentence Prediction and Mutual Information
- SafeConv: Explaining and Correcting Conversational Unsafe Behavior
- Evaluating Open-Domain Question Answering in the Era of Large Language Models
- DiffusionDB: A Large-scale Prompt Gallery Dataset for Text-to-Image Generative Models
- Tell2Design: A Dataset for Language-Guided Floor Plan Generation
- CREPE: Open-Domain Question Answering with False Presuppositions
Student Research Workshop (Oral)
Room: Pier 2&3
- [SRW] Assessing Chain-of-Thought Reasoning against Lexical Negation: A Case Study on Syllogism
- [SRW] Is a Knowledge-based Response Engaging?: An Analysis on Knowledge-Grounded Dialogue with Information Source Annotation
- [SRW] LECO: Improving Early Exiting via Learned Exits and Comparison-based Exiting Mechanism
- [SRW] How-to Guides for Specific Audiences: A Corpus and Initial Findings
Summarization (Oral)
Room: Metropolitan West
- What are the Desired Characteristics of Calibration Sets? Identifying Correlates on Long Form Scientific Summarization
- Balancing Lexical and Semantic Quality in Abstractive Summarization
- Socratic Pretraining: Question-Driven Pretraining for Controllable Summarization
- Attributable and Scalable Opinion Summarization
- Revisiting the Gold Standard: Grounding Summarization Evaluation with Robust Human Evaluation
- [TACL] MACSum: Controllable Summarization with Mixed Attributes
Poster Presentations
Dialogue and Interactive Systems (Poster)
Room: Frontenac Ballroom and Queen's Quay
Discourse and Pragmatics (Poster)
Room: Frontenac Ballroom and Queen's Quay
Information Extraction (Poster)
Room: Frontenac Ballroom and Queen's Quay
Interpretability and Analysis of Models for NLP (Poster)
Room: Frontenac Ballroom and Queen's Quay
Language Grounding to Vision, Robotics, and Beyond (Poster)
Room: Frontenac Ballroom and Queen's Quay
Semantics: Sentence-level Semantics, Textual Inference, and Other Areas (Poster)
Room: Frontenac Ballroom and Queen's Quay
Virtual Poster Presentations
Computational Social Science and Cultural Analytics (Virtual Poster)
Room: Pier 7&8
- Zero-Shot and Few-Shot Stance Detection on Varied Topics via Conditional Generation
- MetaAdapt: Domain Adaptive Few-Shot Misinformation Detection via Meta Learning
- Words as Gatekeepers: Measuring Discipline-specific Terms and Meanings in Scholarly Publications
- Dramatic Conversation Disentanglement
- Counterspeeches up my sleeve! Intent Distribution Learning and Persistent Fusion for Intent-Conditioned Counterspeech Generation
- Race, Gender, and Age Biases in Biomedical Masked Language Models
Dialogue and Interactive Systems (Virtual Poster)
Room: Pier 7&8
- [TACL] Less is More: Mitigate Spurious Correlations for Open-Domain Dialogue Response Generation Models by Causal Discovery
- Pre-training Multi-party Dialogue Models with Latent Discourse Inference
- AutoConv: Automatically Generating Information-seeking Conversations with Large Language Models
- A Probabilistic Framework for Discovering New Intents
- Facilitating Multi-turn Emotional Support Conversation with Positive Emotion Elicitation: A Reinforcement Learning Approach
- Task-Optimized Adapters for an End-to-End Task-Oriented Dialogue System
- Dialogue Planning via Brownian Bridge Stochastic Process for Goal-directed Proactive Dialogue
- Dual Class Knowledge Propagation Network for Multi-label Few-shot Intent Detection
- Bridging The Gap: Entailment Fused-T5 for Open-retrieval Conversational Machine Reading Comprehension
- A Cross-Modality Context Fusion and Semantic Refinement Network for Emotion Recognition in Conversation
- Multimodal Recommendation Dialog with Subjective Preference: A New Challenge and Benchmark
- TREA: Tree-Structure Reasoning Schema for Conversational Recommendation
- SimOAP: Improve Coherence and Consistency in Persona-based Dialogue Generation via Over-sampling and Post-evaluation
- PAL to Lend a Helping Hand: Towards Building an Emotion Adaptive Polite and Empathetic Counseling Conversational Agent
- Imagination is All You Need! Curved Contrastive Learning for Abstract Sequence Modeling Utilized on Long Short-Term Dialogue Planning
- On the Compositional Generalization in Versatile Open-domain Dialogue
- DialoGPS: Dialogue Path Sampling in Continuous Semantic Space for Data Augmentation in Multi-Turn Conversations
- XDailyDialog: A Multilingual Parallel Dialogue Corpus
- On the Correspondence between Compositionality and Imitation in Emergent Neural Communication
- StructSP: Efficient Fine-tuning of Task-Oriented Dialog System by Using Structure-aware Boosting and Grammar Constraints
- DKAF: KB Arbitration for Learning Task-Oriented Dialog Systems with Dialog-KB Inconsistencies
- MMDialog: A Large-scale Multi-turn Dialogue Dataset Towards Multi-modal Open-domain Conversation
Discourse and Pragmatics (Virtual Poster)
Room: Pier 7&8
- Distinguishing Address vs. Reference Mentions of Personal Names in Text
- End-to-End Argument Mining over Varying Rhetorical Structures
- Towards Generative Event Factuality Prediction
- Discourse Analysis via Questions and Answers: Parsing Dependency Structures of Questions Under Discussion
- Connective Prediction for Implicit Discourse Relation Recognition via Knowledge Distillation
Ethics and NLP (Virtual Poster)
Room: Pier 7&8
- Prompt Tuning Pushes Farther, Contrastive Learning Pulls Closer: A Two-Stage Approach to Mitigate Social Biases
- Disagreement Matters: Preserving Label Diversity by Jointly Modeling Item and Annotator Label Distributions with DisCo
- An Empirical Analysis of Parameter-Efficient Methods for Debiasing Pre-Trained Language Models
- MIL-Decoding: Detoxifying Language Models at Token-Level via Multiple Instance Learning
Generation (demo) (Virtual Poster)
Room: Pier 7&8
Generation (Virtual Poster)
Room: Pier 7&8
- Reducing Sensitivity on Speaker Names for Text Generation from Dialogues
- Multi-source Semantic Graph-based Multimodal Sarcasm Explanation Generation
- Context-Aware Document Simplification
- Revisiting the Architectures like Pointer Networks to Efficiently Improve the Next Word Distribution, Summarization Factuality, and Beyond
- Focus-aware Response Generation in Inquiry Conversation
- DiffusionBERT: Improving Generative Masked Language Models with Diffusion Models
- NLG Evaluation Metrics Beyond Correlation Analysis: An Empirical Metric Preference Checklist
- DecompEval: Evaluating Generated Texts as Unsupervised Decomposed Question Answering
- AlignScore: Evaluating Factual Consistency with A Unified Alignment Function
- Tailor: A Soft-Prompt-Based Approach to Attribute-Based Controlled Text Generation
- Unsupervised Graph-Text Mutual Conversion with a Unified Pretrained Language Model
- DivHSK: Diverse Headline Generation using Self-Attention based Keyword Selection
Industry (Virtual Poster)
Room: Pier 7&8
- [Industry] GKD: A General Knowledge Distillation Framework for Large-scale Pre-trained Language Model
- [Industry] Towards Building a Robust Toxicity Predictor
- [Industry] "Knowledge is Power": Constructing Knowledge Graph of Abdominal Organs and Using Them for Automatic Radiology Report Generation
- [Industry] Label efficient semi-supervised conversational intent classification
- [Industry] Tab-CQA: A Tabular Conversational Question Answering Dataset on Financial Reports
- [Industry] Improving Knowledge Production Efficiency With Question Answering on Conversation
- [Industry] Domain-specific transformer models for query translation
- [Industry] Hunt for Buried Treasures: Extracting Unclaimed Embodiments from Patent Specifications
- [Industry] Learn over Past, Evolve for Future: Forecasting Temporal Trends for Fake News Detection
- [Industry] Tab-Cleaner: Weakly Supervised Tabular Data Cleaning via Pre-training for E-commerce Catalog
- [Industry] Boosting Transformers and Language Models for Clinical Prediction in Immunotherapy
- [Industry] Chemical Language Understanding Benchmark
- [Industry] Automated Digitization of Unstructured Medical Prescriptions
- [Industry] Constrained Policy Optimization for Controlled Self-Learning in Conversational AI Systems
- [Industry] CocaCLIP: Exploring Distillation of Fully-Connected Knowledge Interaction Graph for Lightweight Text-Image Retrieval
- [Industry] HyperT5: Towards Compute-Efficient Korean Language Modeling
- [Industry] SPM: A Split-Parsing Method for Joint Multi-Intent Detection and Slot Filling
- [Industry] FashionKLIP: Enhancing E-Commerce Image-Text Retrieval with Fashion Multi-Modal Conceptual Knowledge Graph
- [Industry] SaFER: A Robust and Efficient Framework for Fine-tuning BERT-based Classifier with Noisy Labels
- [Industry] Event-Centric Query Expansion in Web Search
- [Industry] Scalable and Safe Remediation of Defective Actions in Self-Learning Conversational Systems
- [Industry] Content Moderation for Evolving Policies using Binary Question Answering
- [Industry] Building Accurate Low Latency ASR for Streaming Voice Search in E-commerce
- [Industry] Consistent Text Categorization using Data Augmentation in e-Commerce
- [Industry] Transferable and Efficient: Unifying Dynamic Multi-Domain Product Categorization
- [Industry] Rapid Diffusion: Building Domain-Specific Text-to-Image Synthesizers with Fast Inference Speed
Information Extraction (Virtual Poster)
Room: Pier 7&8
- Dual-Gated Fusion with Prefix-Tuning for Multi-Modal Relation Extraction
- Text Augmented Open Knowledge Graph Completion via Pre-Trained Language Models
- An Embarrassingly Easy but Strong Baseline for Nested Named Entity Recognition
- What Is Overlap Knowledge in Event Argument Extraction? APE: A Cross-datasets Transfer Learning Model for EAE
- Multi-hop Evidence Retrieval for Cross-document Relation Extraction
- Graph Propagation based Data Augmentation for Named Entity Recognition
- Easy-to-Hard Learning for Information Extraction
- Type Enhanced BERT for Correcting NER Errors
- Learning Latent Relations for Temporal Knowledge Graph Reasoning
- Silver Syntax Pre-training for Cross-Domain Relation Extraction
- The Art of Prompting: Event Detection based on Type Specific Prompts
- Teamwork Is Not Always Good: An Empirical Study of Classifier Drift in Class-incremental Information Extraction
- Learning with Partial Annotations for Event Detection
- DiSCoMaT: Distantly Supervised Composition Extraction from Tables in Materials Science Articles
- CHEER: Centrality-aware High-order Event Reasoning Network for Document-level Event Causality Identification
- OD-RTE: A One-Stage Object Detection Framework for Relational Triple Extraction
- Semantic Structure Enhanced Event Causality Identification
- Constrained Tuple Extraction with Interaction-Aware Network
- DSP: Discriminative Soft Prompts for Zero-Shot Entity and Relation Extraction
- Adaptive Ordered Information Extraction with Deep Reinforcement Learning
- Learning from a Friend: Improving Event Extraction via Self-Training with Feedback from Abstract Meaning Representation
- Retrieve-and-Sample: Document-level Event Argument Extraction via Hybrid Retrieval Augmentation
- A Novel Table-to-Graph Generation Approach for Document-Level Joint Entity and Relation Extraction
Information Retrieval and Text Mining (Virtual Poster)
Room: Pier 7&8
Interpretability and Analysis of Models for NLP (Virtual Poster)
Room: Pier 7&8
- White-Box Multi-Objective Adversarial Attack on Dialogue Generation
- DSRM: Boost Textual Adversarial Training with Distribution Shift Risk Minimization
- CASN:Class-Aware Score Network for Textual Adversarial Detection
- Conformal Nucleus Sampling
- A Gradient Control Method for Backdoor Attacks on Parameter-Efficient Tuning
- Figurative Language Processing: A Linguistically Informed Feature Analysis of the Behavior of Language Models and Humans
- Hybrid Uncertainty Quantification for Selective Text Classification in Ambiguous Tasks
- Measuring the Instability of Fine-Tuning
- Language Model Analysis for Ontology Subsumption Inference
- ReCode: Robustness Evaluation of Code Generation Models
- Local Interpretation of Transformer Based on Linear Decomposition
- Defending against Insertion-based Textual Backdoor Attacks via Attribution
- Feature Interactions Reveal Linguistic Structure in Language Models
- HuaSLIM: Human Attention Motivated Shortcut Learning Identification and Mitigation for Large Language models
- Dynamic Transformers Provide a False Sense of Efficiency
Language Grounding to Vision, Robotics, and Beyond (Virtual Poster)
Room: Pier 7&8
- Aerial Vision-and-Dialog Navigation
- MultiCapCLIP: Auto-Encoding Prompts for Zero-Shot Multilingual Visual Captioning
- Generate then Select: Open-ended Visual Question Answering Guided by World Knowledge
- Transforming Visual Scene Graphs to Image Captions
- Revealing Single Frame Bias for Video-and-Language Learning
- MS-DETR: Natural Language Video Localization with Sampling Moment-Moment Interaction
- Tackling Modality Heterogeneity with Multi-View Calibration Network for Multimodal Sentiment Detection
- LMCap: Few-shot Multilingual Image Captioning by Retrieval Augmented Language Model Prompting
- Learning from Children: Improving Image-Caption Pretraining via Curriculum
- I Spy a Metaphor: Large Language Models and Diffusion Models Co-Create Visual Metaphors
- A Language-First Approach for Procedure Planning
- Gradient-Boosted Decision Tree for Listwise Context Model in Multimodal Review Helpfulness Prediction
- Transferring General Multimodal Pretrained Models to Text Recognition
Large Language Models (Virtual Poster)
Room: Pier 7&8
- Rethinking Semi-supervised Learning with Language Models
- Leveraging Training Data in Few-Shot Prompting for Numerical Reasoning
- The Larger they are, the Harder they Fail: Language Models do not Recognize Identifier Swaps in Python
- Evaluating the Factual Consistency of Large Language Models Through News Summarization
- A Length-Extrapolatable Transformer
- Plan-and-Solve Prompting: Improving Zero-Shot Chain-of-Thought Reasoning by Large Language Models
- Revisiting Automated Prompting: Are We Actually Doing Better?
- Pre-training Language Model as a Multi-perspective Course Learner
- Scaling Laws for BERT in Low-Resource Settings
- Large Language Models with Controllable Working Memory
- Data-Efficient Finetuning Using Cross-Task Nearest Neighbors
- Towards Adaptive Prefix Tuning for Parameter-Efficient Language Model Fine-tuning
- Multi-target Backdoor Attacks for Code Pre-trained Models
- AraMUS: Pushing the Limits of Data and Model Scale for Arabic Natural Language Processing
- Do Large Language Models Know What They Don't Know?
- Rehearsal-free Continual Language Learning via Efficient Parameter Isolation
Linguistic Diversity (Virtual Poster)
Room: Pier 7&8
Linguistic Theories, Cognitive Modeling, and Psycholinguistics (Virtual Poster)
Room: Pier 7&8
Machine Learning for NLP (Virtual Poster)
Room: Pier 7&8
- Text Augmentation Using Dataset Reconstruction for Low-Resource Classification
- To Copy Rather Than Memorize: A Vertical Learning Paradigm for Knowledge Graph Completion
- LABO: Towards Learning Optimal Label Regularization via Bi-level Optimization
- Structured Pruning for Efficient Generative Pre-trained Language Models
- Self-Distilled Quantization: Achieving High Compression Rates in Transformer-Based Language Models
- Multitask Pre-training of Modular Prompt for Chinese Few-Shot Learning
- MatSci-NLP: Evaluating Scientific Language Models on Materials Science Language Tasks Using Text-to-Schema Modeling
- HiTIN: Hierarchy-aware Tree Isomorphism Network for Hierarchical Text Classification
- PAD-Net: An Efficient Framework for Dynamic Networks
- CAME: Confidence-guided Adaptive Memory Efficient Optimization
- Connectivity Patterns are Task Embeddings
- Low-Rank Updates of pre-trained Weights for Multi-Task Learning
- LaSQuE: Improved Zero-Shot Classification from Explanations Through Quantifier Modeling and Curriculum Learning
- Which Examples Should be Multiply Annotated? Active Learning When Annotators May Disagree
- Know Where You're Going: Meta-Learning for Parameter-Efficient Fine-Tuning
- Distilling Step-by-Step! Outperforming Larger Language Models with Less Training Data and Smaller Model Sizes
- TADA: Efficient Task-Agnostic Domain Adaptation for Transformers
- Exclusive Supermask Subnetwork Training for Continual Learning
- History repeats: Overcoming catastrophic forgetting for event-centric temporal knowledge graph completion
- Grokking of Hierarchical Structure in Vanilla Transformers
- Prototype-Guided Pseudo Labeling for Semi-Supervised Text Classification
- When and how to paraphrase for named entity recognition?
- A Universal Discriminator for Zero-Shot Generalization
- On Dataset Transferability in Active Learning for Transformers
Machine Translation (demo) (Virtual Poster)
Room: Pier 7&8
Machine Translation (Virtual Poster)
Room: Pier 7&8
- Detecting and Mitigating Hallucinations in Machine Translation: Model Internal Workings Alone Do Well, Sentence Similarity Even Better
- A Formal Perspective on Byte-Pair Encoding
- Revisiting Commonsense Reasoning in Machine Translation: Training, Evaluation and Challenge
- What Knowledge Is Needed? Towards Explainable Memory for kNN-MT Domain Adaptation
- Token-Level Self-Evolution Training for Sequence-to-Sequence Learning
- PEIT: Bridging the Modality Gap with Pre-trained Models for End-to-End Image Translation
- MTCue: Learning Zero-Shot Control of Extra-Textual Attributes by Leveraging Unstructured Context in Neural Machine Translation
- Towards Accurate Translation via Semantically Appropriate Application of Lexical Constraints
- Pretrained Bidirectional Distillation for Machine Translation
- Encoder and Decoder, Not One Less for Pre-trained Language Model Sponsored NMT
- How effective is machine translation on low-resource code-switching? A case study comparing human and automatic metrics
- TranSFormer: Slow-Fast Transformer for Machine Translation
- Neural Machine Translation Methods for Translating Text to Sign Language Glosses
Multilingualism and Cross-Lingual NLP (Virtual Poster)
Room: Pier 7&8
- Hybrid Knowledge Transfer for Improved Cross-Lingual Event Detection via Hierarchical Sample Selection
- Mini-Model Adaptation: Efficiently Extending Pretrained Models to New Languages via Aligned Shallow Training
- Zero-shot Cross-lingual Transfer With Learned Projections Using Unlabeled Target-Language Data
- X-RiSAWOZ: High-Quality End-to-End Multilingual Dialogue Datasets and Few-shot Agents
- DLAMA: A Framework for Curating Culturally Diverse Facts for Probing the Knowledge of Pretrained Language Models
- Predicting Human Translation Difficulty Using Automatic Word Alignment
- Unifying Cross-Lingual and Cross-Modal Modeling Towards Weakly Supervised Multilingual Vision-Language Pre-training
- Beyond English-Centric Bitexts for Better Multilingual Language Representation Learning
- On-the-fly Cross-lingual Masking for Multilingual Pre-training
- Parameter-Efficient Finetuning for Robust Continual Multilingual Learning
NLP Applications (demo) (Virtual Poster)
Room: Pier 7&8
NLP Applications (Virtual Poster)
Room: Pier 7&8
- Songs Across Borders: Singable and Controllable Neural Lyric Translation
- Selecting Better Samples from Pre-trained LLMs: A Case Study on Question Generation
- Multilingual Knowledge Graph Completion from Pretrained Language Models with Knowledge Constraints
- Leveraging Prefix Transfer for Multi-Intent Text Revision
- Detecting Adversarial Samples through Sharpness of Loss Landscape
- Prototype-Based Interpretability for Legal Citation Prediction
- KGA: A General Machine Unlearning Framework Based on Knowledge Gap Alignment
- Tucker Decomposition with Frequency Attention for Temporal Knowledge Graph Completion
- Towards Diverse and Effective Question-Answer Pair Generation from Children Storybooks
- Python Code Generation by Asking Clarification Questions
- Dating Greek Papyri with Text Regression
- Sequential Path Signature Networks for Personalised Longitudinal Language Modeling
- Backdooring Neural Code Search
- Bidirectional Transformer Reranker for Grammatical Error Correction
- TransGEC: Improving Grammatical Error Correction with Translationese
- Cross Encoding as Augmentation: Towards Effective Educational Text Classification
- Causality-Guided Multi-Memory Interaction Network for Multivariate Stock Price Movement Prediction
- Learning Non-linguistic Skills without Sacrificing Linguistic Proficiency
- Financial Numeric Extreme Labelling: A dataset and benchmarking
- Commonsense Knowledge Graph Completion Via Contrastive Pretraining and Node Clustering
- AtTGen: Attribute Tree Generation for Real-World Attribute Joint Extraction
- Rethinking Masked Language Modeling for Chinese Spelling Correction
Question Answering (demo) (Virtual Poster)
Room: Pier 7&8
Question Answering (Virtual Poster)
Room: Pier 7&8
- [TACL] Bridging the Gap between Synthetic and Natural Questions via Sentence Decomposition for Semantic Parsing
- Improving Cross-task Generalization of Unified Table-to-text Models with Compositional Task Configurations
- Optimizing Test-Time Query Representations for Dense Retrieval
- Phrase Retrieval for Open Domain Conversational Question Answering with Conversational Dependency Modeling via Contrastive Learning
- S3HQA: A Three-Stage Approach for Multi-hop Text-Table Hybrid Question Answering
- DePlot: One-shot visual language reasoning by plot-to-table translation
- Coupling Large Language Models with Logic Programming for Robust and General Reasoning from Text
- Dynamic Heterogeneous-Graph Reasoning with Language Models and Knowledge Representation Learning for Commonsense Question Answering
- Tab-CoT: Zero-shot Tabular Chain of Thought
- World Models for Math Story Problems
- Reasoning over Hierarchical Question Decomposition Tree for Explainable Question Answering
- When to Read Documents or QA History: On Unified and Selective Open-domain QA
- Distinguish Before Answer: Generating Contrastive Explanation as Knowledge for Commonsense Question Answering
- Multi-Row, Multi-Span Distant Supervision For Table+Text Question Answering
- Product Question Answering in E-Commerce: A Survey
- SkillQG: Learning to Generate Question for Reading Comprehension Assessment
- A Multi-modal Debiasing Model with Dynamical Constraint for Robust Visual Question Answering
- Re-appraising the Schema Linking for Text-to-SQL
Resources and Evaluation (Virtual Poster)
Room: Pier 7&8
- Exploiting Hierarchically Structured Categories in Fine-grained Chinese Named Entity Recognition
- Exploring the Capacity of Pretrained Language Models for Reasoning about Actions and Change
- Personality Understanding of Fictional Characters during Book Reading
- Correction of Errors in Preference Ratings from Automated Metrics for Text Generation
- HeGeL: A Novel Dataset for Geo-Location from Hebrew Text
- RQUGE: Reference-Free Metric for Evaluating Question Generation by Answering the Question
- C-XNLI: Croatian Extension of XNLI Dataset
- ANALOGICAL - A Novel Benchmark for Long Text Analogy Evaluation in Large Language Models
- ORCA: A Challenging Benchmark for Arabic Language Understanding
- FACTUAL: A Benchmark for Faithful and Consistent Textual Scene Graph Parsing
- An Annotated Dataset for Explainable Interpersonal Risk Factors of Mental Disturbance in Social Media Posts
Sentiment Analysis, Stylistic Analysis, and Argument Mining (Virtual Poster)
Room: Pier 7&8
- QAP: A Quantum-Inspired Adaptive-Priority-Learning Model for Multimodal Emotion Recognition
- Making Better Use of Training Corpus: Retrieval-based Aspect Sentiment Triplet Extraction via Label Interpolation
- Span-level Aspect-based Sentiment Analysis via Table Filling
- StoryTrans: Non-Parallel Story Author-Style Transfer with Discourse Representations and Content Enhancing
- AMR-based Network for Aspect-based Sentiment Analysis
- A Unified One-Step Solution for Aspect Sentiment Quad Prediction
- Few-shot Joint Multimodal Aspect-Sentiment Analysis Based on Generative Multimodal Prompt
- Zero-shot Approach to Overcome Perturbation Sensitivity of Prompts
- A Simple Yet Strong Domain-Agnostic De-bias Method for Zero-Shot Sentiment Classification
- TransESC: Smoothing Emotional Support Conversation via Turn-Level State Transition
- A Dataset of Argumentative Dialogues on Scientific Papers
- PAED: Zero-Shot Persona Attribute Extraction in Dialogues
- Balancing the Effect of Training Dataset Distribution of Multiple Styles for Multi-Style Text Transfer
- Measuring Your ASTE Models in The Wild: A Diversified Multi-domain Dataset For Aspect Sentiment Triplet Extraction
- Opinion Tree Parsing for Aspect-based Sentiment Analysis
- MvP: Multi-view Prompting Improves Aspect Sentiment Tuple Prediction
Speech and Multimodality (Virtual Poster)
Room: Pier 7&8
- Prosody-TTS: Improving Prosody with Masked Autoencoder and Conditional Diffusion Model For Expressive Text-to-Speech
- Hybrid Transducer and Attention based Encoder-Decoder Modeling for Speech-to-Text Tasks
- SLUE Phase-2: A Benchmark Suite of Diverse Spoken Language Understanding Tasks
- MOSPC: MOS Prediction Based on Pairwise Comparison
Student Research Workshop (Virtual Poster)
Room: Pier 7&8
- [SRW] Detection and Comparison of Abusive and Hate Speech in English and Hinglish with Emojis Using Deep Learning and Non-Deep Learning Techniques
- [SRW] Towards Efficient Dialogue Processing in the Emergency Response Domain
- [SRW] Authorship Attribution of Late 19th Century Novels using GAN-BERT
- [SRW] Combining Tradition with Modernness: Exploring Event Representations in Vision-and-Language Models for Visual Goal-Step Inference
- [SRW] Moral Mimicry: Large Language Models Produce Moral Rationalizations Tailored to Political Identity
- [SRW] Math Word Problem Solving by Generating Linguistic Variants of Problem Statements
Summarization (Virtual Poster)
Room: Pier 7&8
- CFSum: Coarse-to-Fine Contribution Network for Multimodal Summarization
- Incorporating Distributions of Discourse Structure for Long Document Abstractive Summarization
- Improving Radiology Summarization with Radiograph and Anatomy Prompts
- Towards Argument-Aware Abstractive Summarization of Long Legal Opinions with Summary Reranking
- Multi-Dimensional Evaluation of Text Summarization with In-Context Learning
- Improving Long Dialogue Summarization with Semantic Graph Representation
- Aspect-aware Unsupervised Extractive Opinion Summarization
- Revisiting Cross-Lingual Summarization: A Corpus-based Study and A New Benchmark with Improved Annotation
- Dialogue Summarization with Static-Dynamic Structure Fusion Graph
Syntax: Tagging, Chunking, and Parsing (Virtual Poster)
Room: Pier 7&8
Theme: Reality Check (Virtual Poster)
Room: Pier 7&8
- DITTO: Data-efficient and Fair Targeted Subset Selection for ASR Accent Adaptation
- It Takes Two to Tango: Navigating Conceptualizations of NLP Tasks and Measurements of Performance
- Reimagining Retrieval Augmented Language Models for Answering Queries
- GLUE-X: Evaluating Natural Language Understanding Models from an Out-of-Distribution Generalization Perspective
- Revisiting Non-Autoregressive Translation at Scale
- Reproducibility in NLP: What Have We Learned from the Checklist?
- Towards Reasoning in Large Language Models: A Survey
- Risks and NLP Design: A Case Study on Procedural Document QA
Session 5
⏰ Time: July 11, 16:15-17:45 (EDT) Conference
⏰ Time: (20:15-21:45 UTC)
UTC /
Browser
Oral Presentations
Generation (Oral)
Room: Metropolitan West
- [CL] Neural Data-to-Text Generation Based on Small Datasets: Comparing the Added Value of Two Semi-Supervised Learning Approaches on Top of a Large Language Model
- [TACL] Conditional Generation with a Question-Answering Blueprint
- HAUSER: Towards Holistic and Automatic Evaluation of Simile Generation
- ByGPT5: End-to-End Style-conditioned Poetry Generation with Token-free Language Models
- Are Experts Needed? On Human Evaluation of Counselling Reflection Generation
- Increasing Diversity While Maintaining Accuracy: Text Data Generation with Large Language Models and Human Interventions
Information Extraction (Oral)
Room: Metropolitan Centre
- TAGPRIME: A Unified Framework for Relational Structure Extraction
- Linguistic representations for fewer-shot relation extraction across domains
- Learning Dynamic Contextualised Word Embeddings via Template-based Temporal Adaptation
- MultiTACRED: A Multilingual Version of the TAC Relation Extraction Dataset
- Can NLI Provide Proper Indirect Supervision for Low-resource Biomedical Relation Extraction?
- Open Set Relation Extraction via Unknown-Aware Training
Interpretability and Analysis of Models for NLP (Oral)
Room: Metropolitan East
Linguistic Theories, Cognitive Modeling, and Psycholinguistics (Oral)
Room: Pier 7&8
- [TACL] Why Does Surprisal From Larger Transformer-Based Language Models Provide a Poorer Fit to Human Reading Times?
- Dependency resolution at the syntax-semantics interface: psycholinguistic and computational insights on control dependencies
- Cognitive Reframing of Negative Thoughts through Human-Language Model Interaction
- Exploring How Generative Adversarial Networks Learn Phonological Representations
- A Method for Studying Semantic Construal in Grammatical Constructions with Interpretable Contextual Embedding Spaces
Semantics: Lexical (Oral)
Room: Pier 2&3
- DimonGen: Diversified Generative Commonsense Reasoning for Explaining Concept Relationships
- Does GPT-3 Grasp Metaphors? Identifying Metaphor Mappings with Generative Language Models
- Learning to Substitute Spans towards Improving Compositional Generalization
- LexSym: Compositionality as Lexical Symmetry
- Interpretable Word Sense Representations via Definition Generation: The Case of Semantic Change Analysis
- CLCL: Non-compositional Expression Detection with Contrastive Learning and Curriculum Learning
Poster Presentations
Industry (Poster)
Room: Frontenac Ballroom and Queen's Quay
- [Industry] Federated Learning of Gboard Language Models with Differential Privacy
- [Industry] KG-FLIP: Knowledge-guided Fashion-domain Language-Image Pre-training for E-commerce
- [Industry] AVEN-GR: Attribute Value Extraction and Normalization using product GRaphs
- [Industry] PLAtE: A Large-scale Dataset for List Page Web Extraction
- [Industry] Domain-Agnostic Neural Architecture for Class Incremental Continual Learning in Document Processing Platform
- [Industry] Entity Contrastive Learning in a Large-Scale Virtual Assistant System
- [Industry] An efficient method for Natural Language Querying on Structured Data
- [Industry] Evaluating Embedding APIs for Information Retrieval
- [Industry] AI Coach Assist: An Automated Approach for Call Recommendation in Contact Centers for Agent Coaching
- [Industry] A Static Evaluation of Code Completion by Large Language Models
- [Industry] Predicting Customer Satisfaction with Soft Labels for Ordinal Classification
- [Industry] Application-Agnostic Language Modeling for On-Device ASR
- [Industry] Semantic Ambiguity Detection in Sentence Classification using Task-Specific Embeddings
- [Industry] What, When, and How to Ground: Designing User Persona-Aware Conversational Agents for Engaging Dialogue
- [Industry] Reducing cohort bias in natural language understanding systems with targeted self-training scheme
- [Industry] KoSBI: A Dataset for Mitigating Social Bias Risks Towards Safer Large Language Model Applications
- [Industry] CWSeg: An Efficient and General Approach to Chinese Word Segmentation
- [Industry] Extracting Text Representations for Terms and Phrases in Technical Domains
- [Industry] Weighted Contrastive Learning With False Negative Control to Help Long-tailed Product Classification
- [Industry] RadLing: Towards Efficient Radiology Report Understanding
- [Industry] NAG-NER: a Unified Non-Autoregressive Generation Framework for Various NER Tasks
- [Industry] CUPID: Curriculum Learning Based Real-Time Prediction using Distillation
- [Industry] KAFA: Rethinking Image Ad Understanding with Knowledge-Augmented Feature Adaptation of Vision-Language Models
- [Industry] MathPrompter: Mathematical Reasoning using Large Language Models
- [Industry] Distilled Language Models are economically efficient for the enterprise. ...mostly.
- [Industry] DISCOSQA: A Knowledge Base Question Answering System for Space Debris based on Program Induction
- [Industry] Unified Contextual Query Rewriting
- [Industry] Exploring Zero and Few-shot Techniques for Intent Classification
- [Industry] Generate-then-Retrieve: Intent-Aware FAQ Retrieval in Product Search
- [Industry] EvolveMT: an Ensemble MT Engine Improving Itself with Usage Only
- [Industry] MobileNMT: Enabling Translation in 15MB and 30ms
- [Industry] Multi-doc Hybrid Summarization via Salient Representation Learning
- [Industry] Toward More Accurate and Generalizable Evaluation Metrics for Task-Oriented Dialogs
- [Industry] Mitigating the Burden of Redundant Datasets via Batch-Wise Unique Samples and Frequency-Aware Losses
- [Industry] Search Query Spell Correction with Weak Supervision in E-commerce
- [Industry] Referring to Screen Texts with Voice Assistants
- [Industry] Weakly supervised hierarchical multi-task classification of customer questions
- [Industry] xPQA: Cross-Lingual Product Question Answering in 12 Languages
Machine Learning for NLP (Poster)
Room: Frontenac Ballroom and Queen's Quay
Student Research Workshop (Poster)
Room: Frontenac Ballroom and Queen's Quay
- [SRW] Multi-Dialectal Representation Learning of Sinitic Phonology
- [SRW] Classical Out-of-Distribution Detection Methods Benchmark in Text Classification Tasks
- [SRW] A State-Vector Framework For Dataset Effects
- [SRW] Probing for Hyperbole in Pre-Trained Language Models
- [SRW] Acquiring Frame Element Knowledge with Deep Metric Learning for Semantic Frame Induction
Timezone: Conference (Toronto) UTC Browser
Demo Session 6
⏰ Time: July 12, 09:00-10:30 (EDT) Conference
⏰ Time: (13:00-14:30 UTC)
UTC /
Browser
Poster Presentations
Generation (demo) (Poster)
Room: Frontenac Ballroom and Queen's Quay
Interpretability and Analysis of Models for NLP (demo) (Poster)
Room: Frontenac Ballroom and Queen's Quay
Large Language Models (demo) (Poster)
Room: Frontenac Ballroom and Queen's Quay
Question Answering (demo) (Poster)
Room: Frontenac Ballroom and Queen's Quay
Resources and Evaluation (demo) (Poster)
Room: Frontenac Ballroom and Queen's Quay
Demo Session 7
⏰ Time: July 12, 11:00-12:30 (EDT) Conference
⏰ Time: (15:00-16:30 UTC)
UTC /
Browser
Poster Presentations
Dialogue and Interactive Systems (demo) (Poster)
Room: Frontenac Ballroom and Queen's Quay
Large Language Models (demo) (Poster)
Room: Frontenac Ballroom and Queen's Quay
Machine Learning for NLP (demo) (Poster)
Room: Frontenac Ballroom and Queen's Quay
Poster Session 6
⏰ Time: July 12, 09:00-10:30 (EDT) Conference
⏰ Time: (13:00-14:30 UTC)
UTC /
Browser
Poster Presentations
Computational Social Science and Cultural Analytics (Poster)
Room: Frontenac Ballroom and Queen's Quay
Dialogue and Interactive Systems (Poster)
Room: Frontenac Ballroom and Queen's Quay
- CHBias: Bias Evaluation and Mitigation of Chinese Conversational Language Models
- RECAP: Retrieval-Enhanced Context-Aware Prefix Encoder for Personalized Dialogue Response Generation
- Modeling User Satisfaction Dynamics in Dialogue via Hawkes Process
- Toward Interactive Dictation
- RADE: Reference-Assisted Dialogue Evaluation for Open-Domain Dialogue
- Towards Faithful Dialogues via Focus Learning
- Modeling What-to-ask and How-to-ask for Answer-unaware Conversational Question Generation
- ACCENT: An Automatic Event Commonsense Evaluation Metric for Open-Domain Dialogue Systems
- Prompter: Zero-shot Adaptive Prefixes for Dialogue State Tracking Domain Adaptation
Ethics and NLP (Poster)
Room: Frontenac Ballroom and Queen's Quay
Generation (Poster)
Room: Frontenac Ballroom and Queen's Quay
Information Extraction (Poster)
Room: Frontenac Ballroom and Queen's Quay
- Code4Struct: Code Generation for Few-Shot Event Structure Prediction
- CodeIE: Large Code Generation Models are Better Few-Shot Information Extractors
- Document-Level Multi-Event Extraction with Event Proxy Nodes and Hausdorff Distance Minimization
- Debiasing Generative Named Entity Recognition by Calibrating Sequence Likelihood
- GENEVA: Benchmarking Generalizability for Event Argument Extraction with Hundreds of Event Types and Argument Roles
- AMPERE: AMR-Aware Prefix for Generation-Based Event Argument Extraction Model
Information Retrieval and Text Mining (Poster)
Room: Frontenac Ballroom and Queen's Quay
Interpretability and Analysis of Models for NLP (Poster)
Room: Frontenac Ballroom and Queen's Quay
Language Grounding to Vision, Robotics, and Beyond (Poster)
Room: Frontenac Ballroom and Queen's Quay
- MM-SHAP: A Performance-agnostic Metric for Measuring Multimodal Contributions in Vision and Language Models & Tasks
- Weakly Supervised Vision-and-Language Pre-training with Relative Representations
- End-to-end Knowledge Retrieval with Multi-modal Queries
- Multilingual Conceptual Coverage in Text-to-Image Models
- Modular Visual Question Answering via Code Generation
- Multimodal Relation Extraction with Cross-Modal Retrieval and Synthesis
Large Language Models (Poster)
Room: Frontenac Ballroom and Queen's Quay
- Reasoning with Language Model Prompting: A Survey
- Towards Understanding Chain-of-Thought Prompting: An Empirical Study of What Matters
- Z-ICL: Zero-Shot In-Context Learning with Pseudo-Demonstrations
- Matching Pairs: Attributing Fine-Tuned Models to their Pre-Trained Large Language Models
- Explanation-based Finetuning Makes Models More Robust to Spurious Cues
Linguistic Diversity (Poster)
Room: Frontenac Ballroom and Queen's Quay
Machine Learning for NLP (Poster)
Room: Frontenac Ballroom and Queen's Quay
- Holistic Prediction on a Time-Evolving Attributed Graph
- The Benefits of Bad Advice: Autocontrastive Decoding across Model Layers
- Linear Guardedness and its Implications
- Characterizing and Measuring Linguistic Dataset Drift
- Hidden Schema Networks
- ContraCLM: Contrastive Learning For Causal Language Model
- Is Fine-tuning Needed? Pre-trained Language Models Are Near Perfect for Out-of-Domain Detection
- Learning Neuro-Symbolic World Models with Conversational Proprioception
Machine Translation (Poster)
Room: Frontenac Ballroom and Queen's Quay
- Optimal Transport for Unsupervised Hallucination Detection in Neural Machine Translation
- Memory-efficient NLLB-200: Language-specific Expert Pruning of a Massively Multilingual Machine Translation Model
- Towards Understanding and Improving Knowledge Distillation for Neural Machine Translation
- When Does Translation Require Context? A Data-driven, Multilingual Exploration
Multilingualism and Cross-Lingual NLP (Poster)
Room: Frontenac Ballroom and Queen's Quay
NLP Applications (Poster)
Room: Frontenac Ballroom and Queen's Quay
- DisorBERT: A Double Domain Adaptation Model for Detecting Signs of Mental Disorders in Social Media
- ConFEDE: Contrastive Feature Decomposition for Multimodal Sentiment Analysis
- Trillion Dollar Words: A New Financial Dataset, Task & Market Analysis
- Cross-Modal Attribute Insertions for Assessing the Robustness of Vision-and-Language Learning
- DataFinder: Scientific Dataset Recommendation from Natural Language Descriptions
- Curriculum Learning for Graph Neural Networks: A Multiview Competence-based Approach
- Multitask Pretraining with Structured Knowledge for Text-to-SQL Generation
- Tree-Based Representation and Generation of Natural and Mathematical Language
Phonology, Morphology, and Word Segmentation (Poster)
Room: Frontenac Ballroom and Queen's Quay
Question Answering (Poster)
Room: Frontenac Ballroom and Queen's Quay
Resources and Evaluation (Poster)
Room: Frontenac Ballroom and Queen's Quay
- bgGLUE: A Bulgarian General Language Understanding Evaluation Benchmark
- Can Large Language Models Be an Alternative to Human Evaluations?
- BIG-C: a Multimodal Multi-Purpose Dataset for Bemba
- Evaluating Zero-Shot Event Structures: Recommendations for Automatic Content Extraction (ACE) Annotations
- StoryWars: A Dataset and Instruction Tuning Baselines for Collaborative Story Understanding and Generation
- Is GPT-3 a Good Data Annotator?
Semantics: Lexical (Poster)
Room: Frontenac Ballroom and Queen's Quay
Semantics: Sentence-level Semantics, Textual Inference, and Other Areas (Poster)
Room: Frontenac Ballroom and Queen's Quay
- NatLogAttack: A Framework for Attacking Natural Language Inference Models with Natural Logic
- I2D2: Inductive Knowledge Distillation with NeuroLogic and Self-Imitation
- AMRs Assemble! Learning to Ensemble with Autoregressive Models for AMR Parsing
- LAIT: Efficient Multi-Segment Encoding in Transformers with Layer-Adjustable Interaction
Sentiment Analysis, Stylistic Analysis, and Argument Mining (Poster)
Room: Frontenac Ballroom and Queen's Quay
Speech and Multimodality (Poster)
Room: Frontenac Ballroom and Queen's Quay
Summarization (Poster)
Room: Frontenac Ballroom and Queen's Quay
Theme: Reality Check (Poster)
Room: Frontenac Ballroom and Queen's Quay
Poster Session 7
⏰ Time: July 12, 11:00-12:30 (EDT) Conference
⏰ Time: (15:00-16:30 UTC)
UTC /
Browser
Poster Presentations
Computational Social Science and Cultural Analytics (Poster)
Room: Frontenac Ballroom and Queen's Quay
- Language of Bargaining
- What does a Text Classifier Learn about Morality? An Explainable Method for Cross-Domain Comparison of Moral Rhetoric
- Detoxifying Text with MaRCo: Controllable Revision with Experts and Anti-Experts
- On the Interpretability and Significance of Bias Metrics in Texts: a PMI-based Approach
- Knowledge of cultural moral norms in large language models
Dialogue and Interactive Systems (Poster)
Room: Frontenac Ballroom and Queen's Quay
Discourse and Pragmatics (Poster)
Room: Frontenac Ballroom and Queen's Quay
Ethics and NLP (Poster)
Room: Frontenac Ballroom and Queen's Quay
- Subjective Crowd Disagreements for Subjective Data: Uncovering Meaningful CrowdOpinion with Population-level Learning
- Exploiting Biased Models to De-bias Text: A Gender-Fair Rewriting Model
- BLIND: Bias Removal With No Demographics
- Marked Personas: Using Natural Language Prompts to Measure Stereotypes in Language Models
Generation (Poster)
Room: Frontenac Ballroom and Queen's Quay
Information Extraction (Poster)
Room: Frontenac Ballroom and Queen's Quay
- mOKB6: A Multilingual Open Knowledge Base Completion Benchmark
- WSPAlign: Word Alignment Pre-training via Large-Scale Weakly Supervised Span Prediction
- Revisiting Relation Extraction in the era of Large Language Models
- Actively Supervised Clustering for Open Relation Extraction
- Peeking inside the black box: A Commonsense-aware Generative Framework for Explainable Complaint Detection
Information Retrieval and Text Mining (Poster)
Room: Frontenac Ballroom and Queen's Quay
Interpretability and Analysis of Models for NLP (Poster)
Room: Frontenac Ballroom and Queen's Quay
- Large-Scale Correlation Analysis of Automated Metrics for Topic Models
- An Ordinal Latent Variable Model of Conflict Intensity
- Finding the SWEET Spot: Analysis and Improvement of Adaptive Inference in Low Resource Settings
- Language model acceptability judgements are not always robust to context
- BITE: Textual Backdoor Attacks with Iterative Trigger Injection
- Contrastive Error Attribution for Finetuned Language Models
Language Grounding to Vision, Robotics, and Beyond (Poster)
Room: Frontenac Ballroom and Queen's Quay
Large Language Models (Poster)
Room: Frontenac Ballroom and Queen's Quay
- Dissecting Transformer Length Extrapolation via the Lens of Receptive Field Analysis
- WebCPM: Interactive Web Search for Chinese Long-form Question Answering
- Should you marginalize over possible tokenizations?
- ALERT: Adapt Language Models to Reasoning Tasks
- HINT: Hypernetwork Instruction Tuning for Efficient Zero- and Few-Shot Generalisation
- Token-wise Decomposition of Autoregressive Language Model Hidden States for Analyzing Model Predictions
- Sequence Parallelism: Long Sequence Training from System Perspective
Machine Learning for NLP (Poster)
Room: Frontenac Ballroom and Queen's Quay
- Deriving Language Models from Masked Language Models
- An Invariant Learning Characterization of Controlled Text Generation
- One Network, Many Masks: Towards More Parameter-Efficient Transfer Learning
- In and Out-of-Domain Text Adversarial Robustness via Label Smoothing
- Dataset Distillation with Attention Labels for Fine-tuning BERT
- Improving the robustness of NLI models with minimax training
- Are Message Passing Neural Networks Really Helpful for Knowledge Graph Completion?
Machine Translation (Poster)
Room: Frontenac Ballroom and Queen's Quay
- Tokenization and the Noiseless Channel
- Text Style Transfer Back-Translation
- Exploring Better Text Image Translation with Multimodal Codebook
- WACO: Word-Aligned Contrastive Learning for Speech Translation
- RAMP: Retrieval and Attribute-Marking Enhanced Prompting for Attribute-Controlled Translation
- Scene Graph as Pivoting: Inference-time Image-free Unsupervised Multimodal Machine Translation with Visual Scene Hallucination
Multilingualism and Cross-Lingual NLP (Poster)
Room: Frontenac Ballroom and Queen's Quay
NLP Applications (Poster)
Room: Frontenac Ballroom and Queen's Quay
- Pivotal Role of Language Modeling in Recommender Systems: Enriching Task-specific and Task-agnostic Representation Learning
- Detecting Contradictory COVID-19 Drug Efficacy Claims from Biomedical Literature
- Shrinking Embeddings for Hyper-Relational Knowledge Graphs
- Efficient Diagnosis Assignment Using Unstructured Clinical Notes
- A Compare-and-contrast Multistage Pipeline for Uncovering Financial Signals in Financial Reports
- DT-Solver: Automated Theorem Proving with Dynamic-Tree Sampling Guided by Proof-level Value Function
- U-CREAT: Unsupervised Case Retrieval using Events extrAcTion
- Natural Language to Code Generation in Interactive Data Science Notebooks
- Synthetic Text Generation with Differential Privacy: A Simple and Practical Recipe
- Exploring Continual Learning for Code Generation Models
- VendorLink: An NLP approach for Identifying & Linking Vendor Migrants & Potential Aliases on Darknet Markets
- Compounding Geometric Operations for Knowledge Graph Completion
- Learning Multi-Step Reasoning by Solving Arithmetic Tasks
- Robust Multi-bit Natural Language Watermarking through Invariant Features
Question Answering (Poster)
Room: Frontenac Ballroom and Queen's Quay
- Laziness Is a Virtue When It Comes to Compositionality in Neural Semantic Parsing
- Learning Answer Generation using Supervision from Automatic Question Answering Evaluators
- An Inner Table Retriever for Robust Table Question Answering
- Minding Language Models' (Lack of) Theory of Mind: A Plug-and-Play Multi-Character Belief Tracker
Resources and Evaluation (Poster)
Room: Frontenac Ballroom and Queen's Quay
- FIREBALL: A Dataset of Dungeons and Dragons Actual-Play with Structured Game State Information
- Movie101: A New Movie Understanding Benchmark
- FactKG: Fact Verification via Reasoning on Knowledge Graphs
- IndicMT Eval: A Dataset to Meta-Evaluate Machine Translation Metrics for Indian Languages
- STT4SG-350: A Speech Corpus for All Swiss German Dialect Regions
Semantics: Sentence-level Semantics, Textual Inference, and Other Areas (Poster)
Room: Frontenac Ballroom and Queen's Quay
Sentiment Analysis, Stylistic Analysis, and Argument Mining (Poster)
Room: Frontenac Ballroom and Queen's Quay
Speech and Multimodality (Poster)
Room: Frontenac Ballroom and Queen's Quay
Summarization (Poster)
Room: Frontenac Ballroom and Queen's Quay
Syntax: Tagging, Chunking, and Parsing (Poster)
Room: Frontenac Ballroom and Queen's Quay
Theme: Reality Check (Poster)
Room: Frontenac Ballroom and Queen's Quay
Session 6
⏰ Time: July 12, 09:00-10:30 (EDT) Conference
⏰ Time: (13:00-14:30 UTC)
UTC /
Browser
Oral Presentations
Industry (Oral)
Room: Pier 4&5
- [Industry] Accurate Training of Web-based Question Answering Systems with Feedback from Ranked Users
- [Industry] Reliable and Interpretable Drift Detection in Streams of Short Texts
- [Industry] Answering Unanswered Questions through Semantic Reformulations in Spoken QA
- [Industry] Sharing Encoder Representations across Languages, Domains and Tasks in Large-Scale Spoken Language Understanding
- [Industry] Regression-Free Model Updates for Spoken Language Understanding
- [Industry] "Let's not Quote out of Context": Unified Vision-Language Pretraining for Context Assisted Image Captioning
Machine Learning for NLP (Oral)
Room: Metropolitan Centre
- RankCSE: Unsupervised Sentence Representations Learning via Learning to Rank
- Lifting the Curse of Capacity Gap in Distilling Language Models
- Consistency Regularization Training for Compositional Generalization
- Graph-based Relation Mining for Context-free Out-of-vocabulary Word Embedding Learning
- WhitenedCSE: Whitening-based Contrastive Learning of Sentence Embeddings
Machine Translation (Oral)
Room: Metropolitan West
- Improving Translation Quality Estimation with Bias Mitigation
- Test-time Adaptation for Machine Translation Evaluation by Uncertainty Minimization
- Towards Higher Pareto Frontier in Multilingual Machine Translation
- Causes and Cures for Interference in Multilingual Translation
- Breeding Machine Translations: Evolutionary approach to survive and thrive in the world of automated evaluation
NLP Applications (Oral)
Room: Metropolitan East
- Enhancing Grammatical Error Correction Systems with Explanations
- PMAES: Prompt-mapping Contrastive Learning for Cross-prompt Automated Essay Scoring
- DARE: Towards Robust Text Explanations in Biomedical and Healthcare Applications
- Injecting knowledge into language generation: a case study in auto-charting after-visit care instructions from medical dialogue
- Byte-Level Grammatical Error Correction Using Synthetic and Curated Corpora
- Adaptive and Personalized Exercise Generation for Online Language Learning
Phonology, Morphology, and Word Segmentation (Oral)
Room: Pier 7&8
Semantics: Sentence-level Semantics, Textual Inference, and Other Areas (Oral)
Room: Pier 2&3
- [TACL] Compositional Evaluation on Japanese Textual Entailment and Similarity
- [TACL] Scientia Potentia Est -- On the Role of Knowledge in Computational Argumentation
- Dense-ATOMIC: Towards Densely-connected ATOMIC with High Knowledge Coverage and Massive Multi-hop Paths
- COLA: Contextualized Commonsense Causal Reasoning from the Causal Inference Perspective
- CAT: A Contextualized Conceptualization and Instantiation Framework for Commonsense Reasoning
- [CL] Curing the SICK and other NLI maladies
Poster Presentations
Dialogue and Interactive Systems (Poster)
Room: Frontenac Ballroom and Queen's Quay
Information Retrieval and Text Mining (Poster)
Room: Frontenac Ballroom and Queen's Quay
Linguistic Theories, Cognitive Modeling, and Psycholinguistics (Poster)
Room: Frontenac Ballroom and Queen's Quay
Machine Translation (Poster)
Room: Frontenac Ballroom and Queen's Quay
Phonology, Morphology, and Word Segmentation (Poster)
Room: Frontenac Ballroom and Queen's Quay
Semantics: Lexical (Poster)
Room: Frontenac Ballroom and Queen's Quay
Student Research Workshop (Poster)
Room: Frontenac Ballroom and Queen's Quay
- [SRW] Gender Stereotyping in Popular Children's Videos
- [SRW] The Turing Quest: Can Transformers Make Good NPCs?
- [SRW] Making the Most Out of the Limited Context Length: Predictive Power Varies with Clinical Note Type and Note Section
- [SRW] Intriguing Effect of the Correlation Prior on ICD-9 Code Assignment
- [SRW] Can LMs Store and Retrieve 1-to-N Relational Knowledge?
- [SRW] MedTem2.0: Prompt-based Temporal Classification of Treatment Events from Discharge Summaries
- [SRW] Building a Buzzer-quiz Answering System
- [SRW] Second Language Acquisition of Neural Language Models
- [SRW] Semantic Accuracy in Natural Language Generation: A Thesis Proposal
- [SRW] Towards Reliable Misinformation Mitigation: Generalization, Uncertainty, and GPT-4
Session 7
⏰ Time: July 12, 11:00-12:30 (EDT) Conference
⏰ Time: (15:00-16:30 UTC)
UTC /
Browser
Oral Presentations
Discourse and Pragmatics (Oral)
Room: Pier 2&3
- [TACL] Multilingual Coreference Resolution in Multiparty Dialogue
- [TACL] Coreference Resolution through a seq2seq Transition-Based System
- PairSpanBERT: An Enhanced Language Model for Bridging Resolution
- Annotating Mentions Alone Enables Efficient Domain Adaptation for Coreference Resolution
- Dual Cache for Long Document Neural Coreference Resolution
- Factual or Contextual? Disentangling Error Types in Entity Description Generation
Generation (Oral)
Room: Metropolitan Centre
Information Extraction (Oral)
Room: Metropolitan Centre
Information Retrieval and Text Mining (Oral)
Room: Metropolitan West
- BERM: Training the Balanced and Extractable Representation for Matching to Improve Generalization Ability of Dense Retrieval
- ConvGQR: Generative Query Reformulation for Conversational Search
- Precise Zero-Shot Dense Retrieval without Relevance Labels
- What Are You Token About? Dense Retrieval as Distributions Over the Vocabulary
- FAA: Fine-grained Attention Alignment for Cascade Document Ranking
- CITADEL: Conditional Token Interaction via Dynamic Lexical Routing for Efficient and Effective Multi-Vector Retrieval
Resources and Evaluation (Oral)
Room: Metropolitan East
- WikiHowQA: A Comprehensive Benchmark for Multi-Document Non-Factoid Question Answering
- Toward Human-Like Evaluation for Natural Language Generation with Error Analysis
- SQuARe: A Large-Scale Dataset of Sensitive Questions and Acceptable Responses Created through Human-Machine Collaboration
- QUEST: A Retrieval Dataset of Entity-Seeking Queries with Implicit Set Operations
- A Critical Evaluation of Evaluations for Long-form Question Answering
- Are Human Explanations Always Helpful? Towards Objective Evaluation of Human Natural Language Explanations
Speech and Multimodality (Oral)
Room: Pier 4&5
- Making More of Little Data: Improving Low-Resource Automatic Speech Recognition Using Data Augmentation
- ManagerTower: Aggregating the Insights of Uni-Modal Experts for Vision-Language Representation Learning
- OpenSR: Open-Modality Speech Recognition via Maintaining Multi-Modality Alignment
- Hearing Lips in Noise: Universal Viseme-Phoneme Mapping and Transfer for Robust Audio-Visual Speech Recognition
- Vision Language Pre-training by Contrastive Learning with Cross-Modal Similarity Regulation
- MIR-GAN: Refining Frame-Level Modality-Invariant Representations with Adversarial Network for Audio-Visual Speech Recognition
Poster Presentations
Machine Learning for NLP (Poster)
Room: Frontenac Ballroom and Queen's Quay
Semantics: Sentence-level Semantics, Textual Inference, and Other Areas (Poster)
Room: Frontenac Ballroom and Queen's Quay
Student Research Workshop (Poster)
Room: Frontenac Ballroom and Queen's Quay
- [SRW] How do different tokenizers perform on downstream tasks in scriptio continua languages?: A case study in Japanese
- [SRW] Kanbun-LM: Reading and Translating Classical Chinese in Japanese Methods by Language Models
- [SRW] Aligning Code-Switching Metrics with Bilingual Behavior
- [SRW] Enhancing Ancient Chinese Understanding with Derived Noisy Syntax Trees
- [SRW] Theoretical Linguistics Rivals Embeddings in Language Clustering for Multilingual Named Entity Recognition
- [SRW] EvoGrad: An Online Platform for an Evolving Winograd Schema Challenge using Adversarial Human Perturbations
- [SRW] SWEET: Weakly Supervised Person Name Extraction for Fighting Human Trafficking
Summarization (Poster)
Room: Frontenac Ballroom and Queen's Quay
Virtual Poster Presentations
Computational Social Science and Cultural Analytics (Virtual Poster)
Room: Pier 7&8
- On Text-based Personality Computing: Challenges and Future Directions
- Two Heads Are Better Than One: Improving Fake News Video Detection by Correlating with Neighbors
- Towards Open-Domain Twitter User Profile Inference
- Measuring Intersectional Biases in Historical Documents
- It's not Sexually Suggestive; It's Educative | Separating Sex Education from Suggestive Content on TikTok videos
- Contrastive Learning of Sociopragmatic Meaning in Social Media
- Causal Matching with Text Embeddings: A Case Study in Estimating the Causal Effects of Peer Review Policies
- Responsibility Perspective Transfer for Italian Femicide News
- Supervised Adversarial Contrastive Learning for Emotion Recognition in Conversations
- Ideology Prediction from Scarce and Biased Supervision: Learn to Disregard the "What” and Focus on the "How”!
- Geo-Seq2seq: Twitter User Geolocation on Noisy Data through Sequence to Sequence Learning
- NormMark: A Weakly Supervised Markov Model for Socio-cultural Norm Discovery
Dialogue and Interactive Systems (Virtual Poster)
Room: Pier 7&8
- Multi-Grained Knowledge Retrieval for End-to-End Task-Oriented Dialog
- Model-Based Simulation for Optimising Smart Reply
- Decoupling Pseudo Label Disambiguation and Representation Learning for Generalized Intent Discovery
- Enhancing Personalized Dialogue Generation with Contrastive Latent Variables: Combining Sparse and Dense Persona
- Two Birds One Stone: Dynamic Ensemble for OOD Intent Classification
- Knowledge-enhanced Mixed-initiative Dialogue System for Emotional Support Conversations
- Multi-Domain Dialogue State Tracking with Disentangled Domain-Slot Attention
- How Well Apply Simple MLP to Incomplete Utterance Rewriting?
- How About Kind of Generating Hedges using End-to-End Neural Models?
- DualGATs: Dual Graph Attention Networks for Emotion Recognition in Conversations
- The Whole Truth and Nothing But the Truth: Faithful and Controllable Dialogue Response Generation with Dataflow Transduction and Constrained Decoding
- Learning New Skills after Deployment: Improving open-domain internet-driven dialogue with human feedback
- Injecting Comparison Skills in Task-Oriented Dialogue Systems for Database Search Results Disambiguation
- The CRINGE Loss: Learning what language not to model
- Disfluency Generation for More Robust Dialogue Systems
- RHO (ρ): Reducing Hallucination in Open-domain Dialogues with Knowledge Grounding
Discourse and Pragmatics (Virtual Poster)
Room: Pier 7&8
Ethics and NLP (Virtual Poster)
Room: Pier 7&8
- Gender-tuning: Empowering Fine-tuning for Debiasing Pre-trained Language Models
- FORK: A Bite-Sized Test Set for Probing Culinary Cultural Biases in Commonsense Reasoning Models
- Foveate, Attribute, and Rationalize: Towards Physically Safe and Trustworthy AI
- Social-Group-Agnostic Bias Mitigation via the Stereotype Content Model
- Shielded Representations: Protecting Sensitive Attributes Through Iterative Gradient-Based Projection
- A Comparative Study on the Impact of Model Compression Techniques on Fairness in Language Models
- Causal Intervention for Mitigating Name Bias in Machine Reading Comprehension
- T2IAT: Measuring Valence and Stereotypical Biases in Text-to-Image Generation
- DP-BART for Privatized Text Rewriting under Local Differential Privacy
- Modeling the Q-Diversity in a Min-max Play Game for Robust Optimization
- Modular and On-demand Bias Mitigation with Attribute-Removal Subnetworks
- Debiasing should be Good and Bad: Measuring the Consistency of Debiasing Techniques in Language Models
- Stereotypes and Smut: The (Mis)representation of Non-cisgender Identities by Text-to-Image Models
Generation (Virtual Poster)
Room: Pier 7&8
- MVP: Multi-task Supervised Pre-training for Natural Language Generation
- Teaching the Pre-trained Model to Generate Simple Texts for Text Simplification
- Contrastive Decoding: Open-ended Text Generation as Optimization
- Fantastic Expressions and Where to Find Them: Chinese Simile Generation with Multiple Constraints
- Nano: Nested Human-in-the-Loop Reward Learning for Few-shot Language Model Control
- Differentiable Instruction Optimization for Cross-Task Generalization
- PREADD: Prefix-Adaptive Decoding for Controlled Text Generation
- Efficient Out-of-Domain Detection for Sequence to Sequence Models
- Best-k Search Algorithm for Neural Text Generation
- Few-Shot Data-to-Text Generation via Unified Representation and Multi-Source Learning
- Dynamic and Efficient Inference for Text Generation via BERT Family
- Distilling Reasoning Capabilities into Smaller Language Models
Information Extraction (Virtual Poster)
Room: Pier 7&8
- Enhancing Event Causality Identification with Counterfactual Reasoning
- Uncertainty-Aware Bootstrap Learning for Joint Extraction on Distantly-Supervised Data
- Bootstrapping Neural Relation and Explanation Classifiers
- UTC-IE: A Unified Token-pair Classification Architecture for Information Extraction
- Guide the Many-to-One Assignment: Open Information Extraction via IoU-aware Optimal Transport
- From Ultra-Fine to Fine: Fine-tuning Ultra-Fine Entity Typing Models to Fine-grained
- Data Augmentation for Low-Resource Keyphrase Generation
- Zero- and Few-Shot Event Detection via Prompt-Based Meta Learning
- Early Discovery of Disappearing Entities in Microblogs
- An AMR-based Link Prediction Approach for Document-level Event Argument Extraction
- Document-Level Event Argument Extraction With a Chain Reasoning Paradigm
- Joint Document-Level Event Extraction via Token-Token Bidirectional Event Completed Graph
- CoAug: Combining Augmentation of Labels and Labelling Rules
- QueryForm: A Simple Zero-shot Form Entity Query Framework
- ECG-QALM: Entity-Controlled Synthetic Text Generation using Contextual Q&A for NER
Information Retrieval and Text Mining (Virtual Poster)
Room: Pier 7&8
- Recurrent Attention Networks for Long-text Modeling
- DynaMiTE: Discovering Explosive Topic Evolutions with User Guidance
- Large Language Models are Built-in Autoregressive Search Engines
- Nonparametric Decoding for Generative Retrieval
- SamToNe: Improving Contrastive Loss for Dual Encoder Retrieval Models with Same Tower Negatives
- SimLM: Pre-training with Representation Bottleneck for Dense Passage Retrieval
- Robust Representation Learning with Reliable Pseudo-labels Generation via Self-Adaptive Optimal Transport for Short Text Clustering
- Multiview Identifiers Enhanced Generative Retrieval
- TOME: A Two-stage Approach for Model-based Retrieval
Interpretability and Analysis of Models for NLP (Virtual Poster)
Room: Pier 7&8
- A Hierarchical Explanation Generation Method Based on Feature Interaction Detection
- Emergent Modularity in Pre-trained Transformers
- Counterfactuals of Counterfactuals: a back-translation-inspired approach to analyse counterfactual editors
- Fighting Bias With Bias: Promoting Model Robustness by Amplifying Dataset Biases
- A Close Look into the Calibration of Pre-trained Language Models
- Robustness of Learning from Task Instructions
- SenteCon: Leveraging Lexicons to Learn Human-Interpretable Language Representations
- Adversarial Robustness of Prompt-based Few-Shot Learning for Natural Language Understanding
- NOTABLE: Transferable Backdoor Attacks Against Prompt-based NLP Models
- Don't Retrain, Just Rewrite: Countering Adversarial Perturbations by Rewriting Text
- Contrastive Learning with Adversarial Examples for Alleviating Pathology of Language Model
- Claim-Dissector: An Interpretable Fact-Checking System with Joint Re-ranking and Veracity Prediction
- Instruction Induction: From Few Examples to Natural Language Task Descriptions
- Fine-tuning Happens in Tiny Subspaces: Exploring Intrinsic Task-specific Subspaces of Pre-trained Language Models
- PromptAttack: Probing Dialogue State Trackers with Adversarial Prompts
- Reinforcement Learning for Topic Models
Language Grounding to Vision, Robotics, and Beyond (Virtual Poster)
Room: Pier 7&8
- PuMer: Pruning and Merging Tokens for Efficient Vision Language Models
- Generating Hashtags for Short-form Videos with Guided Signals
- Generating Structured Pseudo Labels for Noise-resistant Zero-shot Video Sentence Localization
- Modularized Zero-shot VQA with Pre-trained Models
- A Multi-Modal Context Reasoning Approach for Conditional Inference on Joint Textual and Visual Clues
- UniFine: A Unified and Fine-grained Approach for Zero-shot Vision-Language Understanding
- Prompt Tuning for Unified Multimodal Pretrained Models
Large Language Models (Virtual Poster)
Room: Pier 7&8
- From Characters to Words: Hierarchical Pre-trained Language Model for Open-vocabulary Language Understanding
- Beyond Positive Scaling: How Negation Impacts Scaling Trends of Language Models
- How Do In-Context Examples Affect Compositional Generalization?
- Nonparametric Masked Language Modeling
- Parameter-efficient Weight Ensembling Facilitates Task-level Knowledge Transfer
- Mixture-of-Domain-Adapters: Decoupling and Injecting Domain Knowledge to Pre-trained Language Models' Memories
- Controlling the Extraction of Memorized Data from Large Language Models via Prompt-Tuning
- Z-Code++: A Pre-trained Language Model Optimized for Abstractive Summarization
- Recipes for Sequential Pre-training of Multilingual Encoder and Seq2Seq Models
- Residual Prompt Tuning: improving prompt tuning with residual reparameterization
- Better Zero-Shot Reasoning with Self-Adaptive Prompting
- Let Me Check the Examples: Enhancing Demonstration Learning via Explicit Imitation
- Masked Latent Semantic Modeling: an Efficient Pre-training Alternative to Masked Language Modeling
- Knowledgeable Parameter Efficient Tuning Network for Commonsense Question Answering
- Improved Logical Reasoning of Language Models via Differentiable Symbolic Programming
- Reasoning in Large Language Models Through Symbolic Math Word Problems
- Honey, I Shrunk the Language: Language Model Behavior at Reduced Scale.
- Gradient Ascent Post-training Enhances Language Model Generalization
Linguistic Diversity (demo) (Virtual Poster)
Room: Pier 7&8
Linguistic Diversity (Virtual Poster)
Room: Pier 7&8
Linguistic Theories, Cognitive Modeling, and Psycholinguistics (Virtual Poster)
Room: Pier 7&8
- Conjunct Lengths in English, Dependency Length Minimization, and Dependency Structure of Coordination
- Automatic Readability Assessment for Closely Related Languages
- Unsupervised Semantic Variation Prediction using the Distribution of Sibling Embeddings
- LMs stand their Ground: Investigating the Effect of Embodiment in Figurative Language Interpretation by Language Models
- Language acquisition: do children and language models follow similar learning stages?
- UniCoRN: Unified Cognitive Signal ReconstructioN bridging cognitive signals and human language
Machine Learning for NLP (demo) (Virtual Poster)
Room: Pier 7&8
Machine Learning for NLP (Virtual Poster)
Room: Pier 7&8
- [CL] Certified Robustness to Text Adversarial Attacks by Randomized [MASK]
- Improving Gradient Trade-offs between Tasks in Multi-task Text Classification
- Reinforced Active Learning for Low-Resource, Domain-Specific, Multi-Label Text Classification
- AD-KD: Attribution-Driven Knowledge Distillation for Language Model Compression
- ECOLA: Enhancing Temporal Knowledge Embeddings with Contextualized Language Representations
- What Makes Pre-trained Language Models Better Zero-shot Learners?
- On the Expressivity Role of LayerNorm in Transformers' Attention
- Peer-Label Assisted Hierarchical Text Classification
- Unsupervised Open-domain Keyphrase Generation
- CFL: Causally Fair Language Models Through Token-level Attribute Controlled Generation
- Exploring Robust Overfitting for Pre-trained Language Models
- Gradient-based Intra-attention Pruning on Pre-trained Language Models
- Real-World Compositional Generalization with Disentangled Sequence-to-Sequence Learning
- Teaching Small Language Models to Reason
- B2T Connection: Serving Stability and Performance in Deep Transformers
- Dialog-Post: Multi-Level Self-Supervised Objectives and Hierarchical Model for Dialogue Post-Training
- Contrastive Bootstrapping for Label Refinement
- Using Domain Knowledge to Guide Dialog Structure Induction via Neural Probabilistic Soft Logic
- Decoder Tuning: Efficient Language Understanding as Decoding
- Free Lunch for Efficient Textual Commonsense Integration in Language Models
- Fourier Transformer: Fast Long Range Modeling by Removing Sequence Redundancy with FFT Operator
- Cost-effective Distillation of Large Language Models
- Efficient Document Embeddings via Self-Contrastive Bregman Divergence Learning
- Semi-Supervised Domain Adaptation for Emotion-Related Tasks
- Multilingual Pre-training with Self-supervision from Global Co-occurrence Information
- Learning to Initialize: Can Meta Learning Improve Cross-task Generalization in Prompt Tuning?
Machine Translation (Virtual Poster)
Room: Pier 7&8
- Fixing MoE Over-Fitting on Low-Resource Languages in Multilingual Machine Translation
- In-context Examples Selection for Machine Translation
- Disambiguated Lexically Constrained Neural Machine Translation
- Learning Optimal Policy for Simultaneous Machine Translation via Binary Search
- Easy Guided Decoding in Providing Suggestions for Interactive Machine Translation
- CMOT: Cross-modal Mixup via Optimal Transport for Speech Translation
- Extract and Attend: Improving Entity Translation in Neural Machine Translation
- Lego-MT: Learning Detachable Models for Massively Multilingual Machine Translation
- Duplex Diffusion Models Improve Speech-to-Speech Translation
- CKDST: Comprehensively and Effectively Distill Knowledge from Machine Translation to End-to-End Speech Translation
- Robustness of Multi-Source MT to Transcription Errors
- CTC-based Non-autoregressive Speech Translation
- Implicit Memory Transformer for Computationally Efficient Simultaneous Speech Translation
Multilingualism and Cross-Lingual NLP (Virtual Poster)
Room: Pier 7&8
- Language Agnostic Multilingual Information Retrieval with Contrastive Learning
- Can Cross-Lingual Transferability of Multilingual Transformers Be Activated Without End-Task Data?
- Language Anisotropic Cross-Lingual Model Editing
- Why Does Zero-Shot Cross-Lingual Generation Fail? An Explanation and a Solution
- Multi-VALUE: A Framework for Cross-Dialectal English NLP
- Cross-Lingual Retrieval Augmented Prompt for Low-Resource Languages
- Tokenization Impacts Multilingual Language Modeling: Assessing Vocabulary Allocation and Overlap Across Languages
- Code-Switched Text Synthesis in Unseen Language Pairs
- Exploring Anisotropy and Outliers in Multilingual Language Models for Cross-Lingual Semantic Sentence Similarity
- Frustratingly Easy Label Projection for Cross-lingual Transfer
- Adversarial Training for Low-Resource Disfluency Correction
- CoLaDa: A Collaborative Label Denoising Framework for Cross-lingual Named Entity Recognition
- Enhancing Few-shot Cross-lingual Transfer with Target Language Peculiar Examples
- Improving Pretraining Techniques for Code-Switched NLP
NLP Applications (Virtual Poster)
Room: Pier 7&8
- MixPAVE: Mix-Prompt Tuning for Few-shot Product Attribute Value Extraction
- Disentangled Phonetic Representation for Chinese Spelling Correction
- Are You Copying My Model? Protecting the Copyright of Large Language Models for EaaS via Backdoor Watermark
- Better Language Models of Code through Self-Improvement
- GEC-DePenD: Non-Autoregressive Grammatical Error Correction with Decoupled Permutation and Decoding
- Zero-Shot Text Classification via Self-Supervised Tuning
- Causal Intervention and Counterfactual Reasoning for Multi-modal Fake News Detection
- An Exploration of Encoder-Decoder Approaches to Multi-Label Classification for Legal and Biomedical Text
- Prompt- and Trait Relation-aware Cross-prompt Essay Trait Scoring
- Towards Identifying Fine-Grained Depression Symptoms from Memes
- Are Pre-trained Language Models Useful for Model Ensemble in Chinese Grammatical Error Correction?
- Improving Grammatical Error Correction with Multimodal Feature Integration
- Contrastive Training Improves Zero-Shot Classification of Semi-structured Documents
- Explainable Recommendation with Personalized Review Retrieval and Aspect Learning
- Distractor Generation based on Text2Text Language Models with Pseudo Kullback-Leibler Divergence Regulation
- MolXPT: Wrapping Molecules with Text for Generative Pre-training
- Scientific Fact-Checking: A Survey of Resources and Approaches
- HermEs: Interactive Spreadsheet Formula Prediction via Hierarchical Formulet Expansion
- Exploring and Verbalizing Academic Ideas by Concept Co-occurrence
- GVdoc - Graph-based Visual DOcument Classification
- AMR-TST: Abstract Meaning Representation-based Text Style Transfer
Phonology, Morphology, and Word Segmentation (Virtual Poster)
Room: Pier 7&8
Question Answering (Virtual Poster)
Room: Pier 7&8
- Counterfactual Multihop QA: A Cause-Effect Approach for Reducing Disconnected Reasoning
- Long-Tailed Question Answering in an Open World
- AttenWalker: Unsupervised Long-Document Question Answering via Attention-based Graph Walking
- TimelineQA: A Benchmark for Question Answering over Timelines
- MVP-Tuning: Multi-View Knowledge Retrieval with Prompt Tuning for Commonsense Reasoning
- Solving Math Word Problems via Cooperative Reasoning induced Language Models
- SConE: Simplified Cone Embeddings with Symbolic Operators for Complex Logical Queries
- KoRC: Knowledge Oriented Reading Comprehension Benchmark for Deep Text Understanding
- Multi-granularity Temporal Question Answering over Knowledge Graphs
- IM-TQA: A Chinese Table Question Answering Dataset with Implicit and Multi-type Table Structures
- Synthesize, Prompt and Transfer: Zero-shot Conversational Question Generation with Pre-trained Language Model
- Hypothetical Training for Robust Machine Reading Comprehension of Tabular Context
- RobustQA: Benchmarking the Robustness of Domain Adaptation for Open-Domain Question Answering
- Knowing-how & Knowing-that: A New Task for Machine Comprehension of User Manuals
Resources and Evaluation (Virtual Poster)
Room: Pier 7&8
- NusaCrowd: Open Source Initiative for Indonesian NLP Resources
- Can Language Models Make Fun? A Case Study in Chinese Comical Crosstalk
- Revisiting Sample Size Determination in Natural Language Understanding
- MedNgage: A Dataset for Understanding Engagement in Patient-Nurse Conversations
- PLUE: Language Understanding Evaluation Benchmark for Privacy Policies in English
- The State of Profanity Obfuscation in Natural Language Processing Scientific Publications
- A Diverse Set of Freely Available Linguistic Resources for Turkish
- A Unified Evaluation Framework for Novelty Detection and Accommodation in NLP with an Instantiation in Authorship Attribution
- LEDA: a Large-Organization Email-Based Decision-Dialogue-Act Analysis Dataset
- ScoNe: Benchmarking Negation Reasoning in Language Models With Fine-Tuning and In-Context Learning
- Take a Break in the Middle: Investigating Subgoals towards Hierarchical Script Generation
- InfoMetIC: An Informative Metric for Reference-free Image Caption Evaluation
- An Inclusive Notion of Text
- Few-shot Adaptation Works with UnpredicTable Data
- NewsMet : A 'do it all' Dataset of Contemporary Metaphors in News Headlines
- K-UniMorph: Korean Universal Morphology and its Feature Schema
- A New Task and Dataset on Detecting Attacks on Human Rights Defenders
- Comparative evaluation of boundary-relaxed annotation for Entity Linking performance
- HaVQA: A Dataset for Visual Question Answering and Multimodal Research in Hausa Language
Semantics: Lexical (Virtual Poster)
Room: Pier 7&8
- A Self-Supervised Integration Method of Pretrained Language Models and Word Definitions
- Together We Make Sense--Learning Meta-Sense Embeddings
- Taxonomy of Problems in Lexical Semantics
- Improving Diachronic Word Sense Induction with a Nonparametric Bayesian method
- DMLM: Descriptive Masked Language Modeling
Semantics: Sentence-level Semantics, Textual Inference, and Other Areas (Virtual Poster)
Room: Pier 7&8
- [TACL] MENLI: Robust Evaluation Metrics from Natural Language Inference
- Ranking-Enhanced Unsupervised Sentence Representation Learning
- Align-then-Enhance: Multilingual Entailment Graph Enhancement with Soft Predicate Alignment
- PIP: Parse-Instructed Prefix for Syntactically Controlled Paraphrase Generation
Sentiment Analysis, Stylistic Analysis, and Argument Mining (Virtual Poster)
Room: Pier 7&8
- USSA: A Unified Table Filling Scheme for Structured Sentiment Analysis
- Don't Lose Yourself! Empathetic Response Generation via Explicit Self-Other Awareness
- Exploiting Rich Textual User-Product Context for Improving Personalized Sentiment Analysis
- DiaASQ: A Benchmark of Conversational Aspect-based Sentiment Quadruple Analysis
Speech and Multimodality (Virtual Poster)
Room: Pier 7&8
- AV-TranSpeech: Audio-Visual Robust Speech-to-Speech Translation
- Listen, Decipher and Sign: Toward Unsupervised Speech-to-Sign Language Recognition
- Joint Speech Transcription and Translation: Pseudo-Labeling with Out-of-Distribution Data
- Zero-shot Visual Question Answering with Language Model Feedback
- Masked Audio Text Encoders are Effective Multi-Modal Rescorers
- Modality Adaption or Regularization? A Case Study on End-to-End Speech Translation
Student Research Workshop (Virtual Poster)
Room: Pier 7&8
Summarization (Virtual Poster)
Room: Pier 7&8
- RISE: Leveraging Retrieval Techniques for Summarization Evaluation
- Summary-Oriented Vision Modeling for Multimodal Abstractive Summarization
- OpineSum: Entailment-based self-training for abstractive opinion summarization
- MeetingBank: A Benchmark Dataset for Meeting Summarization
- An Investigation of Evaluation Methods in Automatic Medical Note Generation
- Towards Unifying Multi-Lingual and Cross-Lingual Summarization
- Generating User-Engaging News Headlines
Syntax: Tagging, Chunking, and Parsing (Virtual Poster)
Room: Pier 7&8
- XSemPLR: Cross-Lingual Semantic Parsing in Multiple Natural Languages and Meaning Representations
- Pre-Trained Language-Meaning Models for Multilingual Parsing and Generation
- Enhancing Unsupervised Semantic Parsing with Distributed Contextual Representations
- A Pilot Study on Dialogue-Level Dependency Parsing for Chinese
- Importance of Synthesizing High-quality Data for Text-to-SQL Parsing
Theme: Reality Check (Virtual Poster)
Room: Pier 7&8
- The Devil is in the Details: On the Pitfalls of Event Extraction Evaluation
- Is Anisotropy Truly Harmful? A Case Study on Text Clustering
- A Call for Standardization and Validation of Text Style Transfer Evaluation
- This prompt is measuring <mask>: evaluating bias evaluation in language models
- Numeric Magnitude Comparison Effects in Large Language Models
- Few-shot Fine-tuning vs. In-context Learning: A Fair Comparison and Evaluation
- GUMSum: Multi-Genre Data and Evaluation for English Abstractive Summarization
- Towards Benchmarking and Improving the Temporal Reasoning Capability of Large Language Models
- Controlling Learned Effects to Reduce Spurious Correlations in Text Classifiers
- How do humans perceive adversarial text? A reality check on the validity and naturalness of word-based adversarial attacks
- Moving Beyond Downstream Task Accuracy for Information Retrieval Benchmarking
Timezone: Conference (Toronto) UTC Browser
⏰ Time: July 13, 09:00-17:00 (EDT) Conference
⏰ Time: (13:00-21:00 UTC)
UTC /
Browser
W10 - The 3rd Workshop on Document-grounded Dialogue and Conversational Question Answering (DialDoc)
⏰ Time: July 13, 09:00-17:00 (EDT) Conference
⏰ Time: (13:00-21:00 UTC)
UTC /
Browser
⏰ Time: July 13, 09:00-17:00 (EDT) Conference
⏰ Time: (13:00-21:00 UTC)
UTC /
Browser
⏰ Time: July 13, 09:00-17:00 (EDT) Conference
⏰ Time: (13:00-21:00 UTC)
UTC /
Browser
⏰ Time: July 13, 09:00-17:00 (EDT) Conference
⏰ Time: (13:00-21:00 UTC)
UTC /
Browser
⏰ Time: July 13, 09:00-17:00 (EDT) Conference
⏰ Time: (13:00-21:00 UTC)
UTC /
Browser
⏰ Time: July 13, 09:00-17:00 (EDT) Conference
⏰ Time: (13:00-21:00 UTC)
UTC /
Browser
⏰ Time: July 13, 09:00-17:00 (EDT) Conference
⏰ Time: (13:00-21:00 UTC)
UTC /
Browser
⏰ Time: July 13, 09:00-17:00 (EDT) Conference
⏰ Time: (13:00-21:00 UTC)
UTC /
Browser
⏰ Time: July 13, 09:00-17:00 (EDT) Conference
⏰ Time: (13:00-21:00 UTC)
UTC /
Browser
⏰ Time: July 13, 09:00-17:00 (EDT) Conference
⏰ Time: (13:00-21:00 UTC)
UTC /
Browser
⏰ Time: July 13, 09:00-17:00 (EDT) Conference
⏰ Time: (13:00-21:00 UTC)
UTC /
Browser
Timezone: Conference (Toronto) UTC Browser
⏰ Time: July 14, 09:00-17:00 (EDT) Conference
⏰ Time: (13:00-21:00 UTC)
UTC /
Browser
⏰ Time: July 14, 09:00-17:00 (EDT) Conference
⏰ Time: (13:00-21:00 UTC)
UTC /
Browser
⏰ Time: July 14, 09:00-17:00 (EDT) Conference
⏰ Time: (13:00-21:00 UTC)
UTC /
Browser
⏰ Time: July 14, 09:00-17:00 (EDT) Conference
⏰ Time: (13:00-21:00 UTC)
UTC /
Browser
⏰ Time: July 14, 09:00-17:00 (EDT) Conference
⏰ Time: (13:00-21:00 UTC)
UTC /
Browser
⏰ Time: July 14, 09:00-17:00 (EDT) Conference
⏰ Time: (13:00-21:00 UTC)
UTC /
Browser
⏰ Time: July 14, 09:00-17:00 (EDT) Conference
⏰ Time: (13:00-21:00 UTC)
UTC /
Browser
⏰ Time: July 14, 09:00-17:00 (EDT) Conference
⏰ Time: (13:00-21:00 UTC)
UTC /
Browser
⏰ Time: July 14, 09:00-17:00 (EDT) Conference
⏰ Time: (13:00-21:00 UTC)
UTC /
Browser