Bio

I am Bokang Zhang, a final-year PhD candidate in the Computer and Information Engineering Department at The Chinese University of Hong Kong, Shenzhen (CUHKSZ), supervised by Prof. Junfeng Wu and Prof. Zhikun Zhang. I am currently a visiting scholar at Emory University, working with Prof. Li Xiong. Before that, I obtained my bachelor's degree in Software Engineering from Zhejiang University in 2021.

My research interests lie in machine learning privacy and security, including differential privacy, privacy-preserving federated learning, and the emerging privacy and security challenges of large language models (LLMs) and LLM-based agents.

I am actively seeking full-time research opportunities in academia and industry and would welcome any relevant discussions.

Research Areas and Papers

• ML Privacy

  • [arXiv'25] FedSGT: Exact Federated Unlearning via Sequential Group-based Training
  • [CCS'24] S2NeRF: Privacy-preserving Training Framework for NeRF

• Differential Privacy

  • [arXiv'25] Bilateral Differentially Private Vertical Federated Boosted Decision Trees

• ML Security

  • [TIFS'25] Online Reward Poisoning in Reinforcement Learning With Convergence Guarantee
  • [CoRL-SAFE-ROL] Exposing Vulnerabilities in RL: A Novel Stealthy Backdoor Attack through Reward Poisoning
  • [arXiv'24] Online Poisoning Attack Against Reinforcement Learning under Black-box Environments