Mechanistic Data Attribution: Tracing the Training Origins of Interpretable LLM Units
Accepted by International Conference on Machine Learning (ICML), Oral Presentation (Top 0.7%), 2026
Recommended citation: Chen, Jianhui, Yuzhang Luo, and Liangming Pan. "Mechanistic Data Attribution: Tracing the Training Origins of Interpretable LLM Units." arXiv preprint arXiv:2601.21996 (2026).
Download Paper