Look Before You Leap: Autonomous Exploration for LLM Agents
For LLM agent developers, this work formalizes autonomous exploration as a key capability and provides a training method, but the results are qualitative and lack quantitative benchmarks.
The paper identifies that LLM agents fail in unfamiliar environments due to premature exploitation, and proposes the Explore-then-Act paradigm with a training strategy interleaving task and exploration rollouts. Results show that systematic exploration improves downstream task performance, though no concrete numbers are provided.
Large language model based agents often fail in unfamiliar environments due to premature exploitation: a tendency to act on prior knowledge before acquiring sufficient environment-specific information. We identify autonomous exploration as a critical yet underexplored capability for building adaptive agents. To formalize and quantify this capability, we introduce Exploration Checkpoint Coverage, a verifiable metric that measures how broadly an agent discovers key states, objects, and affordances. Our systematic evaluation reveals that agents trained with standard task-oriented reinforcement learning consistently exhibit narrow and repetitive behaviors that impede downstream performance. To address this limitation, we develop a training strategy that interleaves task-execution rollouts and exploration rollouts, with each type of rollout optimized by its corresponding verifiable reward. Building on this training strategy, we propose the Explore-then-Act paradigm, which decouples information-gathering from task execution: agents first utilize an interaction budget to acquire grounded environmental knowledge, then leverage it for task resolution. Our results demonstrate that learning to systematically explore is imperative for building generalizable and real-world-ready agents.