报告题目：LLM Security and Safety: Taxonomy, Current, and Future 

报告人：郭文博，美国加州大学教授

会议时间： 2024年9月4日（周三）15:00-17:00

会议地点：清华大学 FIT 3-230

Abstract：

In this talk, I will introduce my understanding of the taxonomy of LLM security and safety and its status. I will zoom in and discuss our recent work in using reinforcement learning for large language model jailbreaking. Finally, I will zoom out again to discuss potential future endeavors for making LLMs more secure and safe. 

Bio：

Wenbo Guo is an assistant professor and Zhu chair at the UCSB CS department. His research lies in the intersection of trustworthy machine learning and ML for computer security. His recent research includes designing generative models, foundation models, and DRL agents for vulnerability analysis and mitigation, as well as improving the explainability and robustness of code generative models, LLMs, and DRL. He is a recipient of the IBM Ph.D. Fellowship (2020-2022), Facebook/Baidu Ph.D. Fellowship Finalist (2020), and ACM CCS Outstanding Paper Award (2018). His research has been featured by multiple mainstream media and has appeared in a diverse set of top-tier venues in security and machine learning. Going beyond academic research, he also actively participates in many world-class cybersecurity competitions, including the AIxCC competition as part of the Shellphish team.