Stop Wandering, Find the Keys: LLMs Discriminate Key States for Efficient Multi-Agent Exploration

ArXi:2410.02511v2 Announce Type: replace With expansive state-action spaces, efficient multi-agent exploration remains a longstanding challenge in reinforcement learning. Although pursuing novelty, diversity, or uncertainty attracts increasing attention, redundant efforts brought by exploration without proper guidance choices poses a practical issue for the community. This paper