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689 results for “keyword:"mechanistic interpretability"” in ICLR 2027. 142 withdrawn or desk rejected listed apart.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]ProbeFool: Evading LLM Jailbreak Activation Monitors with Probe-Agnostic Attention Routing▼4.3k2.00
[PDF]Sparse Autoencoders vs. Transcoders: Encoder Direction Use and Single-Feature Discrimination▼1361.33
[PDF]Small Transformers Learn Kepler's Equation with a Number Line and an Eccentricity Register▼5.8k2.33
[PDF]Causal-Role Recurrence, Geometric Variation, and Objective-Conditioned Readout Across Mamba-1 Scales▲11.00
[PDF]Looped Transformers under the Jacobian Lens: Does the Global Workspace Survive Recurrence?▼4.3k2.00
[PDF]CircuitFACT: Circuit Discovery for Mechanistic Interpretability of Graph Neural Networks▲3.0k2.33
[PDF]NAMED FROM THE START: CONCEPT-ANCHORED SPARSE AUTOENCODERS FOR INTERPRETABILITY OF VISION-LANGUAGE MODELS▲11k2.00
[PDF]Linear Lobotomy For Alien Graph Representations : Probing, Ablation and Recovery of Linear Features in Graph Neural Networks▼4.3k2.00
[PDF]Same State, Different Decisions: Disentangling Formation and Context-Conditioned Execution Mechanisms in VLMs▲3952.67
[PDF]Who Is Left of Whom? Tracing Spatial Evidence and Role Binding in Relative-Position Reasoning▲6.1k2.00
[PDF]CKA's Blind Spot: Why Representation Analysis Misses Policy Change in RLVR, and How to Fix It▼1.3k1.75
[PDF]Where Function Goes: Functional Correspondence by Partial Coupling Across Mechanistic Levels▼4.3k2.00
[PDF]FactGRASP: Graph-Aware Relational Activation Shaping for Factual Knowledge Discovery, Recall, and Transfer in LLMs▲6.1k2.00
[PDF]Personalized but Unchecked: Safe Personalization Requires Conditional Control in Medical Advice▼1.2k1.67
[PDF]What Does a Hallucination Score Tell Us? Auditing the Evidence from Prediction to Mechanistic Claims▼4.5k2.67
[PDF]Beyond Retraining Stability: Auditing Duplication Dependence in Vision Sparse Autoencoders▼12k1.67
Withdrawn or desk rejected (142)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