Chenhe Gu | 顾晨禾

I'm an engineer on foundation model team at Li Auto, where I work on post-training and safety alignment for LLMs and MLLMs. Previously, I was a master's student at UC Irvine, working with Prof. Yao Qin and Dr. Jindong Gu on safety issues and alignment of LLMs and MLLMs. Before that, I obtained my bachelor's degree from Southeast University, Nanjing, China.

Email  /  CV  /  Google Scholar  /  Twitter  /  Github  /  LinkedIn

Research

I'm interested in AI Safety, Robustness, Trustworthy AI, Alignment and Post-training. Representative papers are highlighted.

Crafting highly transferable adversarial examples in Multimodal Large Language Models by incorporating diverse vision-language alignment and perturbing the attention component with Gaussian kernel.