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689 results for “keyword:"mechanistic interpretability"” in ICLR 2027. 142 withdrawn or desk rejected listed apart.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]Weight-Dictionary Decomposition: Reading a Transformer's Residual Stream with Its Own Writes▼1361.33
[PDF]Constructive State Abstraction and Finite Identification in Autoregressive Transformers▼1361.33
[PDF]HealingAxis: Large Language Models Can Innately TherapyMaxxx Clinical Psychotherapeutic Interventions▼1.2k1.67
[PDF]See it, Say it, Sorted: Mechanistic Diagnosis and Parameter-Space Mitigation of Emergent Misalignment in LLMs▲3.0k2.33
[PDF]Low-Rank Dynamics-Effective Latent Carriers for Counterfactual Rollout in Learned World Models▼4.3k2.00
[PDF]A Unified Theoretical Framework of Sparse Dictionary Learning in Mechanistic Interpretability▼1.2k1.67
[PDF]ARGORA: Orchestrated Argumentation with Counterfactual Diagnostics for Multi-LLM Consensus▲9.2k2.67
[PDF]SightLine: What a Vision-Language Model Reads From an Image Is Not What It Answers From▲5.4k3.00
[PDF]What Role Does Explicit Chain-of-Thought Play in Safety Moderation? A Circuit-Level Study▼4.3k2.00
[PDF]Reading Decoder Trajectories: Training-Free Counterfactual Query-Trajectory Reliability for Small-Object Detection▼5.8k2.33
[PDF]Structural Text Recipients Route Same-Prompt Visual Realization in Diffusion Transformers▼1.2k1.67
Withdrawn or desk rejected (142)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