@inproceedings{
guo2025skewact,
title={SkewAct: Red Teaming Large Language Models via Activation-Skewed Adversarial Prompt Optimization},
author={Hanxi Guo and Siyuan Cheng and Guanhong Tao and Guangyu Shen and ZHUO ZHANG and Shengwei An and Kaiyuan Zhang and Xiangyu Zhang},
booktitle={Red Teaming GenAI: What Can We Learn from Adversaries?},
year={2025},
url={https://openreview.net/forum?id=QPdVqbcBgi}
}