Hi there, welcome! I’m currently a third-year Ph.D. student in Gaoling School of Artificial Intelligence, Renmin University of China. It’s a great fortunate to be advised by Prof. Yong Liu. I earned my B.E. degree in Software Engineering from Huazhong University of Science and Technology in June 2022.

My previous research focused on trustworthy AI: TMLR24, ACL24, CVPR23, arXiv:2410.06851. Currently, I am deeply engaged in the theoretical analysis of superalignment, especially weak-to-strong generalization: arXiv:2502.01458, arXiv:2502.11107.

Preprints

* indicates equal contribution

Revisiting Weak-to-Strong Generalization in Theory and Practice: Reverse KL vs. Forward KL

Wei Yao*, Wenkai Yang*, Ziqiao Wang, Yankai Lin, Yong Liu
arXiv preprint arXiv:2502.11107

Understanding the Capabilities and Limitations of Weak-to-Strong Generalization

Wei Yao*, Wenkai Yang*, Ziqiao Wang, Yankai Lin, Yong Liu
arXiv preprint arXiv:2502.01458

Understanding Model Ensemble in Transferable Adversarial Attack

Wei Yao*, Zeliang Zhang*, Huayi Tang, Yong Liu
arXiv preprint arXiv:2410.06851

Selected Publications

* indicates equal contribution

Understanding Fairness Surrogate Functions in Algorithmic Fairness

Wei Yao*, Zhanke Zhou*, Zhicong Li, Bo Han, Yong Liu
TMLR 2024 (TMLR-to-ICLR pilot program)

Revisiting Pre-training Period of Large Language Models

Chen Qian*, Jie Zhang*, Wei Yao*, Dongrui Liu, Zhenfei Yin, Yu Qiao, Yong Liu, Jing Shao
ACL 2024

Random Smooth-based Certified Defense against Text Adversarial Attack

Zeliang Zhang*, Wei Yao*, Susan Liang, Chenliang Xu
EACL 2024

Fair Scratch Tickets: Finding Fair Sparse Networks without Weight Training

Pengwei Tang*, Wei Yao*, Zhicong Li, Yong Liu
CVPR 2023

Research Intern

• 2023.10-2024.3: Shanghai AI Laboratory, research intern, mentor: Dr. Jing Shao.

Honors and Awards

National Scholarship, 2019

Service

Conference Reviewer: ICLR 2025, AISTATS 2025

Journal Reviewer: TMLR