2 papers
cs.CR2026
Don't Trust Your Upstream: Exploiting LLM Multi-Agent System via Topology-Guided Adversarial Propagation
Ruichao Liang, Le Yin, Jing Chen +6
The digital world is witnessing the rapid rise of LLM-based multi-agent systems (MASs) and their powerful applications. However, their security remains insufficiently understood, a…
cs.LG2024
Privacy-preserving Universal Adversarial Defense for Black-box Models
Qiao Li, Cong Wu, Jing Chen +6
Deep neural networks (DNNs) are increasingly used in critical applications such as identity authentication and autonomous driving, where robustness against adversarial attacks is c…