Yuzhe Gu   顾宇喆

PhD Student

OpenMMLab, Shanghai AI Laboratory
School of Electronic Information and Electrical Engineering,
Shanghai Jiao Tong University

Email; Google Scholar; Github

Biography

I am a PhD student at Shanghai Jiao Tong University, in the joint program at Shanghai AI Laboratory, advised by Wenwei Zhang and Kai Chen. Before that, I received the bachelor degree at Wuhan University.

My research interests lie primarily in the area of Large Language Model (LLM). Currently, I mainly work on Hallucination in LLM, focusing on improving its faithfulness. I also have experience in Fine-Tuning (SFT & RLHF), AI Security (Backdoor Attack & Defence), and System Security (Side-Channel Attack & Denfence).

Discussions and cooperations are welcomed!

News

Publications

* denotes equal contribution.

(Co-) First author Papers

ANAH-v2: Scaling Analytical Hallucination Annotation of Large Language Models
Yuzhe Gu*, Ziwei Ji*, Wenwei Zhang, Chengqi Lyu, Dahua Lin, Kai Chen
arXiv 2024, Under Review
[Paper] [Code] [Project]
ANAH: Analytical Annotation of Hallucinations in Large Language Models
Ziwei Ji*, Yuzhe Gu*, Wenwei Zhang, Chengqi Lyu, Dahua Lin, Kai Chen
The 62nd Annual Meeting of the Association for Computational Linguistics (ACL) , 2024
[Paper] [Code] [Project]
One more set: Mitigating conflict-based cache side-channel attacks by extending cache set
Yuzhe Gu, Ming Tang, Quancheng Wang, Han Wang, Haili Ding
Journal of Systems Architecture (JSA)
[Paper]

Co-author Papers

BackCache: Mitigating contention-based cache timing attacks by hiding cache line evictions
Quancheng Wang, Xige Zhang, Han Wang, Yuzhe Gu, Ming Tang
arXiv 2023, Under Review
[Paper]
Redeem myself: Purifying backdoors in deep learning models using self attention distillation
Xueluan Gong, Yanjiao Chen, Wang Yang, Qian Wang, Yuzhe Gu, Huayang Huang, Chao Shen
44th IEEE Symposium on Security and Privacy (Oakland), 2023
[Paper]

Awards