paper
referenced-only
2024
paper:bhalla-towards-unifying-interpretability-and-co-2024Towards unifying interpretability and control: Evaluation via intervention
ByUsha Bhalla·Suraj Srinivas·Asma Ghandeharioun·Himabindu Lakkaraju
Related work— refs + corpus + external arXiv
Cited / in-corpus / arXiv badges show which signals surfaced each row. Multi-source rows weighted higher.
- Locate, Steer, and Improve: A Practical Survey of Actionable Mechanistic Interpretability in Large Language ModelsZhihao Zhang, Mingyang Wang, Zunhai Su, Yiwei Wang, Qianli Wang, Shuzhou Yuan, Ercong Nie, Xufeng Duan, Feijiang Han, Qibo Xue, Zeping Yu, Chenming Shang, Xiao Liang, Jing Xiong, Hui Shen, Chaofan Tao, Zhengwu Liu, Senjie Jin, Zhiheng Xi, Dongdong Zhang, Sophia Ananiadou, Tao Gui, Ruobing Xie, Hayden Kwok-Hay So, Hinrich Sch\"utze, Xuanjing Huang, Qi Zhang, Ngai Wong Hengyuan Zhang2026≈ 76%
- Interpretability as Alignment: Making Internal Understanding a Design PrinciplePratinav Seth, Vinay Kumar Sankarapu Aadit Sengupta2025≈ 75%
- ≈ 75%
- From Attribution to Action: A Human-Centered Application of Activation SteeringMaximilian Dreyer, Katharina Weitz, Wojciech Samek, Sebastian Lapuschkin Tobias Labarta2026≈ 74%
- Interpreting Agentic Systems: Beyond Model Explanations to System-Level AccountabilityDhari Gandhi, Himanshu Joshi, Ahmad Rezaie Mianroodi, Sedef Akinli Kocak, Dhanesh Ramachandran Judy Zhu2026≈ 74%
- Towards Unified Attribution in Explainable AI, Data-Centric AI, and Mechanistic InterpretabilityTessa Han, Usha Bhalla, Himabindu Lakkaraju Shichang Zhang2025≈ 74%
- Position: Mechanistic Interpretability Must Disclose Identification Assumptions for Causal ClaimsFengming Liu Zezheng Lin2026≈ 73%
- Interpretability without actionability: mechanistic methods cannot correct language model errors despite near-perfect internal representationsSadiq Y. Patel, Parth Sheth, Bhairavi Muralidharan, Namrata Elamaran, Aakriti Kinra, John Morgan, Rajaie Batniji Sanjay Basu2026≈ 72%
- Verbalized Eval Awareness Inflates Measured Safetyin corpus2026≈ 72%
- Beyond Transcription: Mechanistic Interpretability in ASRYael Segal-Feldman, Hilit Segev, Aviv Shamsian, Asaf Buchnick, Gill Hetz, Ethan Fetaya, Joseph Keshet, Aviv Navon Neta Glazer2025≈ 72%
- When Should an AI Act? A Human-Centered Model of Scene, Context, and Behavior for Agentic AI DesignDaehoo Yoon, Sung Gyu Koh, Young Hwan Kim, Yehan Ahn, Sung Park Soyoung Jung2026≈ 72%
- Beyond Task Success: An Evidence-Synthesis Framework for Evaluating, Governing, and Orchestrating Agentic AIJoshua Andreas Wellbrock Christopher Koch2026≈ 71%
- Activation Steering via Generative Causal MediationAmir Zur, Atticus Geiger, Dylan Hadfield-Menell Aruna Sankaranarayanan2026≈ 71%
- Mechanistic Interpretability as Statistical Estimation: A Variance AnalysisFran\c{c}ois Portet, Maxime Peyrard Maxime M\'eloux2026≈ 71%
- ≈ 71%
- Overseeing Agents Without Constant Oversight: Challenges and OpportunitiesHussein Mozannar, Maya Murad, Jingya Chen, Saleema Amershi, Adam Fourney Madeleine Grunde-McLaughlin2026≈ 71%
- Interpreting Language Model Parametersin corpus2026≈ 69%
- ≈ 67%
- What Models Express, Suppress, and Resist: Auditing Open-Weight LLMs with Persona Vectorsin corpus2026≈ 67%
- ≈ 67%
- Open Problems in Mechanistic Interpretabilityin corpus2025≈ 67%
- ≈ 67%
- ≈ 67%
- Anatomy of Post-Training: Using Interpretability to Characterize Data and Shape the Learning Signalin corpus2026≈ 67%
- Building and evaluating alignment auditing agentsin corpus2025≈ 66%
- ≈ 66%
- ≈ 66%
- ≈ 66%
Similar preprints — Semantic Scholar
Cited by (2)
- Do Sparse Autoencoders Capture Concept Manifolds?
Sparse autoencoders trained on Llama3.1-8B (layer 19, 500M Pile tokens, 25 SAEs across Standard, JumpReLU, TopK, BatchTopK and Matryoshka architectures at VE > 0.85) do not capture continuous concept
- Manifold Steering Reveals the Shared Geometry of Neural Network Representation and Behavior
Manifold steering — intervening on model activations along paths constrained to lie on a learned activation manifold M_h rather than along Euclidean linear directions — produces behavioral trajectorie