Locate, Steer, and Improve: A Practical Survey of Actionable Mechanistic Interpretability in Large Language Models
Year: 2026
Type: conference-paper
Abstract: Hengyuan Zhang, Zhihao Zhang, Ercong Nie, Mingyang Wang, Zunhai Su, Yiwei Wang, Qianli Wang, Shuzhou Yuan, Xufeng Duan, 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 Schuetze, Xuanjing Huang, Qi Zhang, Ngai Wong. Findings of the Association for Computational Linguistics: ACL 2026. 2026.
Authors: Hengyuan Zhang, Zhihao Zhang, Ercong Nie, Mingyang Wang, Zunhai Su, Yiwei Wang, Qianli Wang, Shuzhou Yuan, Xufeng Duan, 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 Schuetze, Xuanjing Huang, Qi Zhang, Ngai Wong
Institutions: University of Hong Kong, Fudan University, Ludwig-Maximilians-Universität München, Tsinghua University, Technische Universität Darmstadt, Technische Universität Berlin, Technische Universität Dresden, Chinese University of Hong Kong, Nanjing University, University of Manchester, Dartmouth College, University of California, Los Angeles, Tencent (China), Microsoft (United States)
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: hybrid
Awards: G1187148928, G5542150374
OpenAlex ID: W7166846670