Interpretability

What Do Visual Tokens Really Encode? Uncovering Sparsity and Redundancy in Multimodal Large Language Models CVPR 2026 What Makes a Good Query? Measuring the Impact of Human-Confusing Linguistic Features on LLM Performance EACL 2026 What Matters to an LLM? Behavioral and Computational Evidences from Summarization EACL 2026 What NLP Gets Wrong About Contact: Implications for Field Linguistic Evidence EACL 2026 When Do Models Actually Decide? Mapping the Layer-Wise Decision Timeline in Pretrained Neural Networks CVPR 2026 When Meanings Meet: Investigating the Emergence and Quality of Shared Concept Spaces during Multilingual Language Model Training EACL 2026 When Reasoning Collapses: A Depth-Aware Probe into LLM Reasoning (Student Abstract) AAAI 2026 When Seeing Overrides Knowing: Disentangling Knowledge Conflicts in Vision-Language Models ACL 2026 When Smiley Turns Hostile: Interpreting How Emojis Trigger LLMs’ Toxicity AAAI 2026 When Truth Is Overridden: Uncovering the Internal Origins of Sycophancy in Large Language Models AAAI 2026 Where Does Vision Meet Language? Understanding and Refining Visual Fusion in MLLMs via Contrastive Attention CVPR 2026 Where Do LLMs Compose Meaning? A Layerwise Analysis of Compositional Robustness EACL 2026 Where is the Watermark? Interpretable Watermark Detection at the Block Level WACV 2026 Where MLLMs Attend and What They Rely On: Explaining Autoregressive Token Generation CVPR 2026 Where Paths Split: Localized, Calibrated Control of Moral Reasoning in Large Language Models ACL 2026 Why Does Reinforcement Learning Generalize? A Feature-Level Mechanistic Study of Post-Training in Large Language Models ACL 2026 Why LLMs Hallucinate on Structured Knowledge: A Mechanistic Analysis of Reasoning over Linearized Representations ACL 2026 Why Steering Works: Toward a Unified View of Language Model Parameter Dynamics ACL 2026 Word Surprisal Correlates with Sentential Contradiction in LLMs EACL 2026 Workflow Search Reinforcement Learning over Structured Decompositions L4DC 2026 X-MuTeST: A Multilingual Benchmark for Explainable Hate Speech Detection and a Novel LLM-Consulted Explanation Framework AAAI 2026 I^2MoE: Interpretable Multimodal Interaction-aware Mixture-of-Experts ICML 2025 3DGraphLLM: Combining Semantic Graphs and Large Language Models for 3D Scene Understanding ICCV 2025 Ab Initio Nonparametric Variable Selection for Scalable Symbolic Regression with Large p ICML 2025 A Calibrated Reflection Approach for Enhancing Confidence Estimation in LLMs NAACL 2025