EvoPolicyGym：在交互式环境中评估自主策略进化

摘要

自主智能体日益被期望通过反馈来改进可执行策略，然而现有评估往往将这一过程简化为最终得分，或将其与开放式软件工程进展混为一谈。我们提出自主策略进化（Autonomous Policy Evolution）这一受控评估设置，在该设置中，一个框架模型智能体在固定的交互预算下反复编辑可执行策略系统。我们将此设置实例化为EvoPolicyGym基准，该基准基于紧凑的交互式强化学习环境构建，用于评估智能体如何迭代改进探索到的策略。在EvoPolicyGym测试套件上，GPT-5.5在所有16个环境中取得了最强的综合排名得分和排名前两名的表现。除排行榜结果外，EvoPolicyGym还提供轨迹级诊断，用于区分智能体如何分配预算、将反馈转化为参数调整。这些分析表明，强大的自主策略进化不仅依赖于孤立的任务胜利，更取决于发现任务适配的机制，并在受限反馈下精炼策略。

English

Autonomous agents are increasingly expected to improve executable policies through feedback, yet existing evaluations often collapse this process into a final score or confound it with open-ended software-engineering progress. We introduce Autonomous Policy Evolution, a controlled evaluation setting in which a harness-model agent repeatedly edits an executable policy system under a fixed interaction budget. We instantiate this setting in EvoPolicyGym, a benchmark built from compact interactive RL environments that evaluates how agents iteratively improve explored policies. On the EvoPolicyGym suite, GPT-5.5 achieves the strongest aggregate rank score and top-two performance on all 16 environments. Beyond leaderboard results, EvoPolicyGym also provides trajectory-level diagnostics that distinguish how agents allocate budget, convert feedback into parametric tuning. These analyses show that strong autonomous policy evolution depends not only on isolated task wins, but on discovering task-appropriate mechanisms and refining policies under bounded feedback.