NeurIPS 2025 Workshop MechInterp Submissions

ReFIne: A Framework for Trustworthy Large Reasoning Models with Reliability, Faithfulness, and Interpretability
Chung-En Sun, Ge Yan, Akshay R. Kulkarni, Tsui-Wei Weng
- Published: 30 Sept 2025, Last Modified: 15 Oct 2025
- Mech Interp Workshop (NeurIPS 2025) Poster
- Readers: Everyone
Narrow Finetuning Leaves Clearly Readable Traces in the Activation Differences
Julian Minder, Clément Dumas, Stewart Slocum, Neel Nanda
- Published: 30 Sept 2025, Last Modified: 30 Sept 2025
- Mech Interp Workshop (NeurIPS 2025) Spotlight
- Readers: Everyone
Localizing Reasoning Training-Induced Changes in Large Language Models
Max Klabunde, Florian Lemmerich
- Published: 30 Sept 2025, Last Modified: 30 Sept 2025
- Mech Interp Workshop (NeurIPS 2025) Poster
- Readers: Everyone
Rank-1 Reasoning: Minimal Parameter Diffs Encode Interpretable Reasoning Signals
Jake Ward, Paul M. Riechers, Adam Shai
- Published: 30 Sept 2025, Last Modified: 30 Sept 2025
- Mech Interp Workshop (NeurIPS 2025) Poster
- Readers: Everyone
Unsupervised decoding of encoded reasoning using language model interpretability
Ching Fang, Samuel Marks
- Published: 30 Sept 2025, Last Modified: 30 Sept 2025
- Mech Interp Workshop (NeurIPS 2025) Spotlight
- Readers: Everyone
Just-in-time and distributed task representations in language models
Yuxuan Li, Declan Iain Campbell, Stephanie C.Y. Chan, Andrew Kyle Lampinen
- Published: 30 Sept 2025, Last Modified: 16 Oct 2025
- Mech Interp Workshop (NeurIPS 2025) Spotlight
- Readers: Everyone
Faithfulness through Causal Abstraction: Aligning explanations of how models reason
Mette Friis Andersen, Ana Lucic, Maria Heuss
- Published: 30 Sept 2025, Last Modified: 30 Sept 2025
- Mech Interp Workshop (NeurIPS 2025) Poster
- Readers: Everyone
Token Entanglement in Subliminal Learning
Amir Zur, Zhuofan Ying, Alexander Russell Loftus, Kerem Şahin, Steven Yu, Lucia Quirke, Tamar Rott Shaham, Natalie Shapira, Hadas Orgad, David Bau
- Published: 30 Sept 2025, Last Modified: 30 Sept 2025
- Mech Interp Workshop (NeurIPS 2025) Poster
- Readers: Everyone
Eliciting Secret Knowledge from Language Models
Bartosz Cywiński, Emil Ryd, Rowan Wang, Senthooran Rajamanoharan, Neel Nanda, Arthur Conmy, Samuel Marks
- Published: 30 Sept 2025, Last Modified: 09 Oct 2025
- Mech Interp Workshop (NeurIPS 2025) Poster
- Readers: Everyone
Function Induction and Task Generalization: An Interpretability Study with Off-by-One Addition
Qinyuan Ye, Robin Jia, Xiang Ren
- Published: 30 Sept 2025, Last Modified: 30 Sept 2025
- Mech Interp Workshop (NeurIPS 2025) Poster
- Readers: Everyone
Detecting and Characterizing Planning in Language Models
Jatin Nainani, Sankaran Vaidyanathan, Connor Watts, Andre N. Assis, Alice Rigg
- Published: 30 Sept 2025, Last Modified: 30 Sept 2025
- Mech Interp Workshop (NeurIPS 2025) Poster
- Readers: Everyone
Dense SAE Latents Are Features, Not Bugs
Xiaoqing Sun, Alessandro Stolfo, Joshua Engels, Ben Peng Wu, Senthooran Rajamanoharan, Mrinmaya Sachan, Max Tegmark
- Published: 30 Sept 2025, Last Modified: 30 Sept 2025
- Mech Interp Workshop (NeurIPS 2025) Spotlight
- Readers: Everyone
Reverse-Engineering Memory in DreamerV3: From Sparse Representations to Functional Circuits
Jan Sobotka, Auke Ijspeert, Guillaume Bellegarda
- Published: 30 Sept 2025, Last Modified: 21 Oct 2025
- Mech Interp Workshop (NeurIPS 2025) Spotlight
- Readers: Everyone
From Black-box to Causal-box: Towards Building More Interpretable Models
Inwoo Hwang, Yushu Pan, Elias Bareinboim
- Published: 30 Sept 2025, Last Modified: 24 Oct 2025
- Mech Interp Workshop (NeurIPS 2025) Poster
- Readers: Everyone
From Local to Contextually-Enriched Local Representations: A Mechanism for Holistic Processing in DINOv2 ViTs
Fenil R. Doshi, Thomas Fel, Talia Konkle, George A. Alvarez
- Published: 30 Sept 2025, Last Modified: 30 Sept 2025
- Mech Interp Workshop (NeurIPS 2025) Poster
- Readers: Everyone
LLM Probing with Contrastive Eigenproblems: Improving Understanding and Applicability of CCS
Stefan F. Schouten, Peter Bloem
- Published: 30 Sept 2025, Last Modified: 30 Sept 2025
- Mech Interp Workshop (NeurIPS 2025) Poster
- Readers: Everyone
OpenMAIA: a Multimodal Automated Interpretability Agent based on open-source models
Josep Lopez Camuñas, Christy Li, Tamar Rott Shaham, Antonio Torralba, Agata Lapedriza
- Published: 30 Sept 2025, Last Modified: 30 Sept 2025
- Mech Interp Workshop (NeurIPS 2025) Poster
- Readers: Everyone
Understanding sparse autoencoder scaling in the presence of feature manifolds
Eric J Michaud, Liv Gorton, Tom McGrath
- Published: 30 Sept 2025, Last Modified: 30 Sept 2025
- Mech Interp Workshop (NeurIPS 2025) Poster
- Readers: Everyone
From Narrow Unlearning to Emergent Misalignment: Causes, Consequences, and Containment in LLMs
Erum Mushtaq, Anil Ramakrishna, Satyapriya Krishna, Sattvik Sahai, Prasoon Goyal, Kai-Wei Chang, Tao Zhang, Rahul Gupta
- Published: 30 Sept 2025, Last Modified: 30 Sept 2025
- Mech Interp Workshop (NeurIPS 2025) Poster
- Readers: Everyone
Better Hessians Matter: Studying the Impact of Curvature Approximations in Influence Functions
Dat Minh Hong, Bruno Kacper Mlodozeniec, Runa Eschenhagen, Richard E. Turner
- Published: 30 Sept 2025, Last Modified: 30 Sept 2025
- Mech Interp Workshop (NeurIPS 2025) Spotlight
- Readers: Everyone
Adaptive Task Vectors for Large Language Models
Joonseong Kang, Soojeong Lee, Subeen Park, Sumin Park, Taero Kim, Jihee Kim, Ryunyi LEE, Kyungwoo Song
- Published: 30 Sept 2025, Last Modified: 30 Sept 2025
- Mech Interp Workshop (NeurIPS 2025) Poster
- Readers: Everyone
Looking into Black Box Code Language Models
Muhammad Umair Haider, Umar Farooq, A.B. Siddique, Mark Marron
- Published: 30 Sept 2025, Last Modified: 30 Sept 2025
- Mech Interp Workshop (NeurIPS 2025) Poster
- Readers: Everyone
Transformers Don’t Need LayerNorm at Inference Time: Scaling LayerNorm Removal to GPT-2 XL and the Implications for Mechanistic Interpretability
Luca Baroni, Galvin Khara, Joachim Schaeffer, Marat Subkhankulov, Stefan Heimersheim
- Published: 30 Sept 2025, Last Modified: 30 Sept 2025
- Mech Interp Workshop (NeurIPS 2025) Poster
- Readers: Everyone
The Anatomy of Alignment: Decomposing Preference Optimization by Steering Sparse Features
Jeremias Lino Ferrao, Matthijs van der Lende, Ilija Lichkovski, Clement Neo
- Published: 30 Sept 2025, Last Modified: 27 Oct 2025
- Mech Interp Workshop (NeurIPS 2025) Spotlight
- Readers: Everyone
Controlling Vision–Language–Action Policies through Sparse Latent Directions
Momin Ahmad Khan, Novak Boskov, Fatima M. Anwar, Manzoor A. Khan
- Published: 30 Sept 2025, Last Modified: 30 Sept 2025
- Mech Interp Workshop (NeurIPS 2025) Poster
- Readers: Everyone