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689 results for “keyword:"mechanistic interpretability"” in ICLR 2027. 142 withdrawn or desk rejected listed apart.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]Reconstruction, Not Localization: A Mechanistic Account of Sparse-Feature OOD Detection in Vision Transformers▼1361.33
[PDF]Navigating the OCEAN: An Atlas of Big Five Personality Representations in Language Models▲8.9k2.75
[PDF]How Vision-Language Models Form Ambiguous Depth Interpretations: A Sparse Feature Analysis of Visual Evidence Transformation▼1.2k1.67
[PDF]Early Context Starvation in Conversational LLMs: Causal Evidence from Attention Interventions▼1.2k1.67
[PDF]One Token Flips the Verdict: Hypergraph Propagation and State Readout Dissociate in Selective Belief Revision▼5.9k2.50
[PDF]Latent Assembly: Causal Discovery and Cross-Task Linking of Executable Operations in Language Models▲13k2.33
[PDF]Predicting Reasoning Errors Before They Occur: A Prompt-Side Activation Direction Anticipates Correctness in a Large Language Model▼1361.33
[PDF]Steering Language Models Without Contrastive Pairs via Unsupervised Latent Region Matching▼1361.33
[PDF]Aggregation Order in Activation Patching: Which Circuit Is Selected and What It Preserves▼4.3k2.00
[PDF]MechLeash-FL: Mechanistic Leashing of Latent Features for Federated Learning under Feature Skew▲382.80
[PDF]Predictive Mechanistic Interpretability of Post-Training: A Priori Forecasts of Circuit Localization and Sign from the Base Model▼2.0k1.00
[PDF]From Target Selection to Condition Reuse: Auditing Residual Interventions in Tool-Use Models▼4.3k2.00
[PDF]De-Absorption Reparameterization: Post-hoc Repair of Feature Absorption in Sparse Autoencoders.▲3.0k2.33
[PDF]Structure Before Flexibility: Mechanistically Testable Functional Conductances for Continuous-Time Network Dynamics▼5.8k2.33
[PDF]Unstable Features, Reproducible Subspaces: Understanding Seed Dependence in Sparse Autoencoders▼5.8k2.33
Withdrawn or desk rejected (142)
[PDF]Softmax Attention is an Exact Gradient Step: Predictable Finite Responses from RoPE DerivativesWithdrawn2.00
[PDF]Whose Word Survives Compression?Desk rejected
[PDF]Vision-Language Models Can Report Which Image's Visual-Token Activations Were PerturbedWithdrawn1.67
[PDF]Same Tokens, Different Readings: Representational and Causal Tests of Lexical-Unit Encoding in Language ModelsWithdrawn1.00