Publications

RADAR: A Risk-Aware Dynamic Multi-Agent Framework for LLM Safety Evaluation Leveraging Role-Specialized Collaboration

Xiuyuan Chen, Jian Zhao, Yuchen Yuan, Tianle Zhang, Huilin Zhou, Zheng Zhu, Linghe Kong, Chi Zhang, Weiran Huang, Xuelong Li. "RADAR: A Risk-Aware Dynamic Multi-Agent Framework for LLM Safety Evaluation Leveraging Role-Specialized Collaboration."

CodeMimicry: Exploiting Safety Generalization Lag in Large Language Models via Structured Code Completion

Zhen Liang, Jian Zhao, Huilin Zhou, Tianle Zhang, Xuelong Li, Hai Huang. "CodeMimicry: Exploiting Safety Generalization Lag in Large Language Models via Structured Code Completion."

RaGEP: Rank-aware Geometric Expert Pruning for Mixture-of-Experts Language Models

Wentao Hu, Zeyu Zhu, Mingkuan Zhao, Zhenhua An, Yanbo Zhai, Shanhong yu, Huilin Zhou, Xin Lai, Xiaoyan Zhu, Jiayin Wang. "RaGEP: Rank-aware Geometric Expert Pruning for Mixture-of-Experts Language Models."

Disson: Automated Multimodal Jailbreaking via Cross-Modal Cognitive Dissonance

Yuxuan Zhou, Kuofeng Gao, Zhuochen Fan, Huilin Zhou, Zesheng Shi, Yang Bai, Tao Dai, Shu-Tao Xia. "Disson: Automated Multimodal Jailbreaking via Cross-Modal Cognitive Dissonance."

TeleAI-Safety: A comprehensive LLM jailbreaking benchmark towards attacks, defenses, and evaluations

Xiuyuan Chen, Jian Zhao, Yuxiang He, Yuan Xun, Xinwei Liu, Yanshu Li, Huilin Zhou, Wei Cai, Ziyan Shi, Yuchen Yuan, Tianle Zhang, Chi Zhang, Xuelong Li. "TeleAI-Safety: A comprehensive LLM jailbreaking benchmark towards attacks, defenses, and evaluations." arXiv:2512.05485, 2025.

Morpheus: Learning to Jailbreak via Self-Evolving Metacognition

Huilin Zhou, Jian Zhao, YiluZhong, Zhen Liang, Xiuyuan Chen, Tianle Zhang, Yuchen Yuan, Chi Zhang, Lan Zhang, Xuelong Li. "Morpheus: Learning to Jailbreak via Self-Evolving Metacognition."