Papers by Xufeng Zhang
Locate, Steer, and Improve: A Practical Survey of Actionable Mechanistic Interpretability in Large Language Models (2026.findings-acl)
Copied to clipboard
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
| Challenge: | Existing literature on mechanistic interpretation (MI) treats it as an observational science, leaving practical applications underexplored. |
| Approach: | They propose a survey structured around the pipeline to identify and improve MI models. |
| Outcome: | The proposed framework enables tangible improvements in Alignment, Capability, and Efficiency. |
RL-Guider: Leveraging Historical Decisions and Feedback for Drug Editing with Large Language Models (2025.findings-acl)
Copied to clipboard
| Challenge: | RLs can be used to refine drugs by iterative conversations with domain experts . existing methods do not leverage past knowledge, but human experts develop intuition over time through historical experience . |
| Approach: | They propose a reinforcement-learning agent to provide suggestions to large language models . RL-Guider leverages the “world-level” knowledge of LLMs and historical feedback . |
| Outcome: | a new reinforcement-learning agent improves the performance of large language models . the proposed agent leverages the “world-level” knowledge of LLMs and historical feedback . |
Towards Provably Secure Generative AI: Reliable Consensus Sampling (2026.findings-acl)
Copied to clipboard
Yu Cui, Hang Fu, Sicheng Pan, Zhuoyu Sun, Yifei Liu, Yuhong Nie, Bo Ran, Baohan Huang, Xufeng Zhang, Haibin Zhang, Cong Zuo, Licheng Wang
| Challenge: | Existing research on generative AI security is driven by mutually reinforcing attack and defense methodologies grounded in empirical experience. |
| Approach: | They propose a new algorithm that uses a random sampling algorithm to control risk. |
| Outcome: | The proposed algorithm improves robustness and utility while maintaining latency comparable to existing algorithms. |