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
98 results for “keyword:"activation patching"” in ICLR 2027. 28 withdrawn or desk rejected listed apart. clearsort:relevanceratingnewestmost read
[PDF]Predicting Reasoning Errors Before They Occur: A Prompt-Side Activation Direction Anticipates Correctness in a Large Language Model▼1351.33
[PDF]Aggregation Order in Activation Patching: Which Circuit Is Selected and What It Preserves▼4.3k2.00
[PDF]Knowing Whom to Trust: Reliability States and Information Use in Large Language Models▼5.8k2.33
[PDF]Does the Workspace Do the Work? A Causal Protocol for Verifying Global-Workspace Computation in a Language Model▲5103.00
[PDF]Utility-Preserving Relabeling Reveals Semantic Sensitivity in Language-Model Decision Making▼1.2k1.67
[PDF]A Linear Name-Origin Prior in the Recall Heads: How Language Models Fabricate and Misrecall Facts About People▲5103.00
[PDF]Right on Average, Wrong per Respondent: A Mechanistic Diagnosis of LLM-Simulated Choice▼1.2k1.67
[PDF]INTERVENTION IS NOT IDENTIFICATION: ACTIVATION EDITS CAN PRODUCE THE PREDICTED EFFECT FOR THE WRONG REASON▼1351.33
[PDF]Out of Sight, Still in Mind? Separating Internal and Answer-Level Effects of Retained Key-Value History▼1.2k1.67
[PDF]RouteVerify: Routing Interventions for Step-Level Verification in Mixture-of-Experts Models▲3.0k2.33
[PDF]Fireman against Jody: Causal Counterfactual Mirrors for Auditing Strategic Deception in Language Models▼1.2k1.67
[PDF]Where Does the Audio Jailbreak Live? A Controlled Frequency-Depth Audit of Audio-Language Models▼1.2k1.67
[PDF]Representational Inertia: Activation Stitching Reverses Fine-Tuned Behavior in Large Language Models▼4.3k2.00
[PDF]Interventional Commutativity: Effect Preservation and Reference Selection Under Weight-Only Quantisation▼1.2k1.67
[PDF]Small Transformers Learn Kepler's Equation with a Number Line and an Eccentricity Register▼5.8k2.33
[PDF]Routing-Guided Single-Expert Interventions Measured Against a Model’s Own Execution Variability▲6.1k2.00
[PDF]What Do Activation Interventions Establish? Causal Controls for Autoregressive Time-Series Foundation Models▼4.3k2.00
[PDF]Content-Dependent Recovery from Task-Relevant Activation Disruption in Large Language Models▲4052.67
Withdrawn or desk rejected (28)
[PDF]Same Head, Different Calls? Auditing Functional Role Claims in Looped TransformersWithdrawn1.33
[PDF]Token-axis Predictive Decomposition: What Information Does Each Token Add to the Hidden States of a Language Model?Withdrawn1.33
[PDF]Same Answer, Different Futures: Judging Latent States by Their Continuations in Looped TransformersDesk rejected
[PDF]Sparse Attention-Head Patching for Cross-Notation Numerical Comparison: Donor-Position TransferWithdrawn1.33