Natural Language Processing

When Annotators Disagree, Topology Explains: Mapper, a Topological Tool for Exploring Text Embedding Geometry and Ambiguity EMNLP 2025 When Attention Sink Emerges in Language Models: An Empirical View ICLR 2025 When Audio and Text Disagree: Revealing Text Bias in Large Audio-Language Models EMNLP 2025 When Bad Data Leads to Good Models ICML 2025 When Detection Fails: The Power of Fine-Tuned Models to Generate Human-Like Social Media Text ACL 2025 When Does Meaning Backfire? Investigating the Role of AMRs in NLI EMNLP 2025 When Do LLMs Help With Node Classification? A Comprehensive Analysis ICML 2025 When Every Token Counts: Optimal Segmentation for Low-Resource Language Models COLING 2025 When Evolution Strategy Meets Language Models Tuning COLING 2025 When Format Changes Meaning: Investigating Semantic Inconsistency of Large Language Models EMNLP 2025 When Harry Meets Superman: The Role of The Interlocutor in Persona-Based Dialogue Generation ACL 2025 When in Doubt, Ask First: A Unified Retrieval Agent-Based System for Ambiguous and Unanswerable Question Answering IJCNLP 2025 When Inverse Data Outperforms: Exploring the Pitfalls of Mixed Data in Multi-Stage Fine-Tuning EMNLP 2025 When is Task Vector Provably Effective for Model Editing? A Generalization Analysis of Nonlinear Transformers ICLR 2025 When Large Language Models Meet Speech: A Survey on Integration Approaches ACL 2025 When Less Is More: Logits-Constrained Framework with RoBERTa for Ancient Chinese NER NAACL 2025 When Life Gives You Samples: The Benefits of Scaling up Inference Compute for Multilingual LLMs EMNLP 2025 When LLMs Struggle: Reference-less Translation Evaluation for Low-resource Languages COLING 2025 When Long Helps Short: How Context Length in Supervised Fine-tuning Affects Behavior of Large Language Models EMNLP 2025 When Men Bite Dogs: Testing Good-Enough Parsing in Turkish with Humans and Large Language Models NAACL 2025 When Models Lie, We Learn: Multilingual Span-Level Hallucination Detection with PsiloQA EMNLP 2025 When Multilingual Models Compete with Monolingual Domain-Specific Models in Clinical Question Answering NAACL 2025 When natural language is not enough: The limits of in-context learning demonstrations in multilingual reasoning NAACL 2025 When Neutral Summaries Are Not That Neutral: Quantifying Political Neutrality in LLM-Generated News Summaries (Student Abstract) AAAI 2025 When Open-Vocabulary Visual Question Answering Meets Causal Adapter: Benchmark and Approach AAAI 2025