Interpretability

What Did I Do Wrong? Quantifying LLMs’ Sensitivity and Consistency to Prompt Engineering NAACL 2025 What Do VLMs NOTICE? A Mechanistic Interpretability Pipeline for Gaussian-Noise-free Text-Image Corruption and Evaluation NAACL 2025 What Features in Prompts Jailbreak LLMs? Investigating the Mechanisms Behind Attacks EMNLP 2025 What Language Do Non-English-Centric Large Language Models Think in? ACL 2025 What Makes a Good Reasoning Chain? Uncovering Structural Patterns in Long Chain-of-Thought Reasoning EMNLP 2025 What makes an Ensemble (Un) Interpretable? ICML 2025 What Makes Cryptic Crosswords Challenging for LLMs? COLING 2025 What should a neuron aim for? Designing local objective functions based on information theory ICLR 2025 What’s in a prompt? Language models encode literary style in prompt embeddings EMNLP 2025 What's in the Image? A Deep-Dive into the Vision of Vision Language Models CVPR 2025 What's New in My Data? Novelty Exploration via Contrastive Generation ICLR 2025 What’s the Difference? Supporting Users in Identifying the Effects of Prompt and Model Changes Through Token Patterns ACL 2025 What We Talk About When We Talk About LMs: Implicit Paradigm Shifts and the Ship of Language Models NAACL 2025 When Annotators Disagree, Topology Explains: Mapper, a Topological Tool for Exploring Text Embedding Geometry and Ambiguity EMNLP 2025 When Backdoors Speak: Understanding LLM Backdoor Attacks Through Model-Generated Explanations ACL 2025 When Format Changes Meaning: Investigating Semantic Inconsistency of Large Language Models EMNLP 2025 When LLMs Play the Telephone Game: Cultural Attractors as Conceptual Tools to Evaluate LLMs in Multi-turn Settings ICLR 2025 When LRP Diverges from Leave-One-Out in Transformers EMNLP 2025 When Models Lie, We Learn: Multilingual Span-Level Hallucination Detection with PsiloQA EMNLP 2025 When Models Reason in Your Language: Controlling Thinking Language Comes at the Cost of Accuracy EMNLP 2025 When Truthful Representations Flip Under Deceptive Instructions? EMNLP 2025 Where Confabulation Lives: Latent Feature Discovery in LLMs EMNLP 2025 Where Did That Come From? Sentence-Level Error-Tolerant Attribution EMNLP 2025 Where to show Demos in Your Prompt: A Positional Bias of In-Context Learning EMNLP 2025 Where, What, Why: Towards Explainable Driver Attention Prediction ICCV 2025