学 术 报 告
Jiaheng Zhang
Assistant Professor
National University of Singapore
Efficient Zero-Knowledge Proofs and Applications in Large Language Models
高效零知识证明以及在大模型中的应用
11月11日（周二）16:00
玉泉校区逸夫工商楼101会议室
报告简介
In this talk, we explore the emerging field of Zero-Knowledge Machine Learning (ZKML), which combines cryptographic zero-knowledge proofs with machine learning. We discuss how zero-knowledge proofs can be used to ensure the verifiability of model inference without revealing sensitive model parameters. The talk also examines the application of zero-knowledge techniques to Large Language Model (LLM) inference, highlighting both current progress and open challenges. Finally, we touch on broader applications of ZKML beyond inference, outlining future directions for AI safety with the technique of ZKML.
报告人简介
Jiaheng Zhang is an Assistant Professor in the Department of Computer Science at the National University of Singapore (NUS). He received his Ph.D. from UC Berkeley, advised by Prof. Dawn Song, and his Bachelor's degree from Shanghai Jiao Tong University (ACM Honors Class). His research focuses on computer security and privacy, applied cryptography, zero-knowledge proofs, AI security, and blockchain. His work covers both theoretical algorithms and practical applications, with some results deployed in real-world systems. He has published in top conferences such as CRYPTO, IEEE S&P (Oakland), ACM CCS, USENIX Security, and ASPLOS. He received the Facebook Fellowship, Forbes 30 Under 30 Asia, MIT TR 35 in Asia Pacific, the Yunfan Bright Star Award at the World AI Conference, and Robert Brown Promising UYoung Researcher Award by Ministry of Education, Singapore.
