Search papers, labs, and topics across Lattice.
Zhenzhen Huang, Songbo Zhang, and Jiaquan Zhang are with School of Information and Software Engineering, University of Electronic Science and Technology of China, Chengdu, China (email: alley10086@gmail.com; 2024090914002@std.uestc.edu.cn; jiaquanzhang2005@gmail.com). Chaoning Zhang, Haoyu Bian, Yalan Ye, and Yang Yang are with School of Computer Science and Engineering, University of Electronic Science and Technology of China, Chengdu, China (email: chaoningzhang@uestc.edu.cn; haoyubian04@gmail.com; yalanye@uestc.edu.cn; yang.yang@uestc.edu.cn). Chi-lok Andy Tai is with College of Professional and Continuing Education, The Hong Kong Polytechnic University, Hong Kong, China (email: andy.tai@cpce-polyu.edu.hk). Caiyan Qin is with School of Robotics and Advanced Manufacture, Harbin Institute of Technology, Shenzhen, China (email: qincaiyan@hit.edu.cn). Jingjing Qu is with Shanghai Artificial Intelligence Laboratory, Shanghai, China (e-mail:qujingjing@pjlab.org.cn). Heng Tao Shen is with School of Computer Science and Technology, Tongji University, Shanghai, China (e-mail: shenhengtao@tongji.edu.cn).* Corresponding Author
1
1
3
3
LLM-based multi-agent systems are riddled with 20 distinct risk types, from single-agent vulnerabilities to system-level emergent hazards, demanding a unified safety evaluation and monitoring framework.