Interpretability

How Does Cognitive Bias Affect Large Language Models? A Case Study on the Anchoring Effect in Price Negotiation Simulations EMNLP 2025 How Does DPO Reduce Toxicity? A Mechanistic Neuron-Level Analysis EMNLP 2025 (How) Do Language Models Track State? ICML 2025 How Do Large Vision-Language Models See Text in Image? Unveiling the Distinctive Role of OCR Heads EMNLP 2025 How Do LLMs Acquire New Knowledge? A Knowledge Circuits Perspective on Continual Pre-Training ACL 2025 How Far Is Video Generation from World Model: A Physical Law Perspective ICML 2025 How Jailbreak Defenses Work and Ensemble? A Mechanistic Investigation EMNLP 2025 How Persuasive Is Your Context? EMNLP 2025 How Programming Concepts and Neurons Are Shared in Code Language Models ACL 2025 How Real Are Synthetic Therapy Conversations? Evaluating Fidelity in Prolonged Exposure Dialogues EMNLP 2025 How Reliable are Causal Probing Interventions? IJCNLP 2025 How Reliable is Multilingual LLM-as-a-Judge? EMNLP 2025 How to Fine-Tune Safely on a Budget: Model Adaptation Using Minimal Resources EMNLP 2025 How to Generalize the Detection of AI-Generated Text: Confounding Neurons EMNLP 2025 How to Make LLMs Forget: On Reversing In-Context Knowledge Edits NAACL 2025 How to Probe: Simple Yet Effective Techniques for Improving Post-hoc Explanations ICLR 2025 How to safely discard features based on aggregate SHAP values COLT 2025 How Transformers Learn Structured Data: Insights From Hierarchical Filtering ICML 2025 How Well Can Reasoning Models Identify and Recover from Unhelpful Thoughts? EMNLP 2025 How well do LLMs reason over tabular data, really? ACL 2025 How Your Location Relates to Health: Variable Importance and Interpretable Machine Learning for Environmental and Sociodemographic Data AAAI 2025 Human–Agent Teaming for Higher-Order Thinking Augmentation IJCNLP 2025 Human Cognition-Inspired Hierarchical Fuzzy Learning Machine ICML 2025 Human-in-the-loop or AI-in-the-loop? Automate or Collaborate? AAAI 2025 HumT DumT: Measuring and controlling human-like language in LLMs ACL 2025