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
1,261 results for “keyword:"interpretability"” in ICLR 2027. 264 withdrawn or desk rejected listed apart. clearsort:relevanceratingnewestmost read
[PDF]Don't Ignore Them: Scaffold Features Matter for Stable SAE Steering in Diffusion Models▲3952.67
[PDF]Localization Does Not Imply Steerability: A Mechanistic Study of Sign Errors in LLM Mathematics▼1361.33
[PDF]What Goes Up Should Come Down: A Training-Distribution Control for Interpretability Metrics▲5103.00
[PDF]Matryoshka attribution: Learning to attribute language model outputs to representations and weights▲2623.67
[PDF]Distributed Geographic Identity Encoding in Vision-Language Models: From Linear Probing to Sparse Autoencoder Feature Ablation, and Why Post-Hoc Debiasing Fails▲11.00
[PDF]Detecting Unfaithful Chain-of-Thought via Circuit-Guided Internal-External Discrepancy▼4.5k2.25
[PDF]UNIFYING DYNAMICAL SYSTEMS AND GRAPH THEORY TO MECHANISTICALLY UNDERSTAND COMPUTATION IN NEURAL NETWORKS▼4.3k2.00
[PDF]CRANE: Functional Selectivity of Language-Conditioned Neurons in Multilingual Large Language Models▼1.2k1.67
[PDF]Not Which Path, but What Happens Before You Would Stop: What Reasoning Post-Training Changes▼1361.33
[PDF]Who Grades the Edit? Frozen and Re-fit Readouts for Position Interventions in RoPE Models▼1361.33
[PDF]Computation Persists, Routing Degrades: A Mechanistic Analysis of Transformer Length Extrapolation▼4.3k2.00
[PDF]What Lies Within: Diagnostics of Compositional and Sequential Behavior in Long-Context Text Embeddings▼1.2k1.67
[PDF]Replication Is Not Specificity: Negative Controls for Representation-Based Reward-Hacking Detection▼1361.33
[PDF]SAE Interventions Can Be Unreliable: Post-Intervention Recovery of Suppressed Behavior▼4.3k2.00
Withdrawn or desk rejected (264)
[PDF]A Window Is Not a Door: Mapping the Limits and Conditions of Knowledge Editing in Language ModelsDesk rejected2.33