CAGE Challenge 4 / CybORG

Researchers led by UC Santa Cruz integrated LLM agents into the CybORG CAGE 4 multi-agent defence environment and proposed a communication protocol for mixed LLM and RL teams. In their runs an all-RL team scored far better reward than an all-LLM (GPT-4o-mini) team and acted about 104 times faster, though the authors highlight LLM explainability and note the environment was designed for RL agents.

CAGE Challenge 4, run under The Technical Cooperation Program, asks entrants to build five cooperating blue-team agents that defend a segmented fictional military network against randomized red agents while green agents generate legitimate activity. The challenge ran from February to May 2024 in the CybORG simulator, scored by mean reward over 100 randomized 500-step episodes, and the environment remains public.