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
689 results for “keyword:"mechanistic interpretability"” in ICLR 2027. 142 withdrawn or desk rejected listed apart. clearsort:relevanceratingnewestmost read
[PDF]The Representation-Continuation Gap: Low-Rank States vs. Low-Rank Dynamics in Vision Transformers▼1.2k1.67
[PDF]Factual Access States in Diffusion Language Models: From Mechanism to Knowledge Editing and Retrieval▲5103.00
[PDF]Coherent Overlap: Token-Specific Expert Committees in Sparse Mixture-of-Experts Routing▼1361.33
[PDF]A Wrong Turn in Addition: Generation, Routing, and Correction of Arithmetic Errors in LLMs▲5103.00
[PDF]InvExplain: An Agentic Interpretation Framework for SAE Features via Training-Free Diffusion Inversion▼5.8k2.33
[PDF]Frequency, Direction, and Feature Sharing: Dissecting Co-activation in Sparse Autoencoders▲13k2.33
[PDF]When Native Fit Misleads: Evaluating Attention Surrogates under Positional Interventions▼4.3k2.00
[PDF]Ambient Dimension Governs Whether Frozen Single-Cell Embeddings Can Be Edited and Interpreted▼1.2k1.67
[PDF]Recurrent Refinement Moves Candidate-Quality Readability Earlier and Stabilizes Its Coordinates▼1.2k1.67
[PDF]How Much Attribution Mass Lands on Tokens That Cannot Matter? A Faithfulness Benchmark for Transformers▲5103.00
[PDF]What Makes Position Zero Special? A Mechanistic Study of Position Zero Attention Sinks in LLMs▼5.8k2.33
[PDF]When Does a Novel Form Become a Word? Causal Evidence for Reusable Lexical Function in Frozen Language Models▲3952.67
[PDF]Locating, Removing, and Restoring: Auditing the Modular Account of Safety in Instruction-Tuned LLMs▼4.3k2.00
[PDF]Beyond Linear Representation Hypothesis: Non-Linear Activation Steering in Text-to-Image Models▲5.4k3.00
[PDF]Beyond Static Interpretability: Anticipating Post-SFT Mechanisms from Pre-SFT Parameters for Better Tuning▼1.2k3.00
Withdrawn or desk rejected (142)
[PDF]Why Benign Post-Training Induces Safety Gate Collapse: Changes in Harmful Activation Geometry Relative to the Inherited Refusal GateWithdrawn1.67
[PDF]Concept Subspaces Compute Beyond the Logit Lens: A Weights-Only Test for Locating Representations Upstream of ReadoutWithdrawn