NeurIPS 2026 reviews are up: reviews, scores, rebuttals and meta-reviews for 7,877 papers. See the highest rated
acceptodds: the papers of ICLR 2027
73 results for “keyword:"representation analysis"” in ICLR 2027. 21 withdrawn or desk rejected listed apart. clearsort:relevanceratingnewestmost read
[PDF]Form Before Harm: Diagnosing the Semantic Influence of Jailbreak Formats with Activation Steering▼1.2k1.67
[PDF]What Happens to the Goal? Distinct Failures of Policy Alignment in Goal Misgeneralization▲3972.67
[PDF]Don't Fix What Isn't Broken: Calibrated Repair of Frozen Foundation Models under Product Shift▼4.5k2.67
[PDF]What Is a Visual Thought Worth? When Continuous Visual States Help Vision-Language Models Reason▼4.3k2.00
[PDF]What Do INRs Find Difficult? Understanding and Exploiting Coordinate-Level Training Dynamics▼1.2k1.67
[PDF]Zentropy Semantic Configuration Fields: A Cross-Modal Framework for Analyzing Semantic Organization in Learned Representations▼4.5k2.67
[PDF]CKA's Blind Spot: Why Representation Analysis Misses Policy Change in RLVR, and How to Fix It▼1.3k1.75
[PDF]The Hidden Cost of Prompt Formatting: Measuring Representational Sensitivity in Language Models▼1361.33
[PDF]Grounding Is Inherited, Action Is Concentrated: What VLA Finetuning Adds to a Web-Pretrained VLM▼71.25
[PDF]Understanding Robustness to Observation Corruptions through Actor-Critic Information Specialization▲3972.67
[PDF]The Attribution Blind Spot: Layerwise Trajectory Diagnostics for Source Reliance in Retrieval-Augmented Language Models▼4.3k2.00
[PDF]Risk Exposure Does Not Mean Unsafe Action: Evidence-Dynamics Alignment for Agent Safety▼1361.33
Withdrawn or desk rejected (21)
[PDF]Who Gets a Token, and What Does It Carry? Unequal Name Support and Concept Access in Large Language ModelsWithdrawn1.33
[PDF]Probe-Geometry Alignment: Auditing and Suppressing Representational Memorization in Language ModelsWithdrawn1.25
[PDF]Persona Following Is Not Selective Control: The Neutrality Gap in LLM User SimulationDesk rejected1.33