OpenART: Scaling Agent Red Teaming via Open-Ended Environment Evolution
This work proposes the Evolutionary Markov Hypergraph Attack (EMHA), a black-box policy that performs feedback-driven environment evolution by coordinating authorized state transitions without requiring parameter updates, and establishes OpenART as a scalable foundation for studying agent safety in complex, evolving en...
Yunhao Chen, Xin Wang, Yi-Xu Wang et al.
· 0 citations