I am a PhD student in the ASTAPLE Lab at The Hong Kong Polytechnic University, supervised by Prof.Hu Haibo. Previously, I graduated from Jilin University with a bachelor’s degree and from Institute of Information Engineering, Chinese Academy of Sciences with a master’s degree.

My research interest includes trustworthy machine learning and privacy protection, with a focus on privacy and security attacks.

🔥 News

2025.12: 🎉🎉 Our paper United We Defend: Collaborative Membership Inference Defenses in Federated Learning is accepted by USENIX Security 2026.
2025.09: 🎉🎉 Our paper Toward Efficient Inference Attacks: Shadow Model Sharing via Mixture-of-Experts is accepted by NeurIPS 2025.

📝 Publications

United We Defend: Collaborative Membership Inference Defenses in Federated Learning.
Li Bai, Junxu Liu, Sen Zhang, Xinwei Zhang, Qingqing Ye, and Haibo Hu. USENIX Security, 2026.
[Paper] [Code]
Toward Efficient Inference Attacks: Shadow Model Sharing via Mixture-of-Experts.
Li Bai, Qingqing Ye, Xinwei Zhang, Sen Zhang, Zi Liang, Jianliang Xu, Haibo Hu. The Thirty-Ninth Annual Conference on Neural Information Processing Systems (NeurIPS), 2025.
[Paper] [Code]
ProVFL: Property inference attacks against vertical federated learning.
Li Bai, Xinwei Zhang, Sen Zhang, Qingqing Ye,Haibo Hu. IEEE Transactions Information Forensics and Security (TIFS), 2025.
[Paper] [Code]
RMR: A relative membership risk measure for machine learning models.
Li Bai, Haibo Hu, Qingqing Ye, Jianliang Xu, Jin Li, Chengfang Fang, Jie Shi. IEEE Transactions on Dependable and Secure Computing (TDSC), 2025.
[Paper] [Code]
Auditing MLaaS Inference Service Quality without Ground Truth via Mutual Information.
Zhu Jiang, Haibo Hu, Qingqing Ye, Li Bai. IEEE Transactions Information Forensics and Security (TIFS), 2025.
[Paper]
MER-Inspector: Assessing model extraction risks from an attack-agnostic perspective.
Xinwei Zhang, Haibo Hu, Qingqing Ye, Li Bai, Huadi Zheng. Proceedings of the ACM on Web Conference (WWW), 2025.
[Paper] [Code]
A sample-level evaluation and generative framework for model inversion attacks.
Haoyang Li, Li Bai, Qingqing Ye, Haibo Hu, Yaxin Xiao, Huadi Zheng, Jianliang Xu. Proceedings of the Thirty-Ninth AAAI Conference on Artificial Intelligence (AAAI), 2025.
[Paper]
Membership inference attacks and defenses in federated learning: A survey.
Li Bai, Haibo Hu, Qingqing Ye, Haoyang Li, Leixia Wang, Jianliang Xu. ACM Computing Surveys, 2024.
[Paper]

📖 Educations

2022.09 - now, The Hong Kong Polytechnic University, PhD in Electronic and Information Engineering.
2016.09 - 2019.06, Chinese Academy of Sciences, Institute of Information Engineering, Master in Cyberspace Security.
2012.09 - 2016.06, Jilin University, Bachelor in Information and Network Security.

🎖 Honors and Awards

2017.09 Outstanding Student Awards, University of Chinese Academy of Sciences.
2015.12 First Prize Scholarship, Jilin University (top 5%).
2015.05 Suzhou Industrial Park Scholarship, Jilin University.
2014.12 China National Scholarship, Jilin University (top 2%).
2013.12 China National Scholarship, Jilin University (top 2%).

💻 Experience

2021.05 - 2022.08, Research Assistant, The Hong Kong Polytechnic University, Hong Kong.
2019.07 - 2020.08, Algorithm Engineer, Sogou Inc, Beijing China.

📋 Services

Journal Reviewer

IEEE Transactions on Information Forensics and Security
IEEE Transactions on Dependable and Secure Computing
IEEE Transactions on Computational Social Systems

Teaching Assistant

Information Technology, The Hong Kong Polytechnic University, 2025
Artificial Intelligence and Science Fiction, The Hong Kong Polytechnic University, 2024,
Information Technology, Fundamentals of Machine Intelligence, The Hong Kong Polytechnic University, 2023
Mobile Systems and Application Development, The Hong Kong Polytechnic University, 2022

BAI Li