Better, Faster, Stronger: Programmatic Skill Learning Best Reduces Agent Cost

更快、更强、更优：程序化技能学习是降低智能体成本的最佳途径

Recently, the practice of augmenting LLM agent capability with skills has gained prevalence. We explore the cost effective adaptation of agents to novel domains by means of learning skills. Existing works focus on performance gain over cost effectiveness. As a result, little is known about what skill learning strategies save cost.

最近，通过技能增强大语言模型（LLM）智能体能力的做法日益普及。我们探讨了通过学习技能使智能体以低成本适应新领域的方法。现有的研究多关注性能提升而非成本效益，因此，关于哪些技能学习策略能够有效节省成本，目前所知甚少。

We argue that among all the different skill learning methods, those that view skills as programs can achieve the best cost reduction. By executing sequences of actions deterministically, a program-augmented agent can reliably and cheaply achieve goals that would otherwise require trial and error and risk degenerate behavior over long horizons.

我们认为，在所有不同的技能学习方法中，将技能视为“程序”的方法能够实现最佳的成本削减。通过确定性地执行动作序列，程序增强型智能体可以可靠且低成本地实现目标，而这些目标在其他情况下往往需要反复试错，且在长周期任务中存在行为退化的风险。

An agent can learn at inference time by incrementally discovering these programs and equipping them for future tasks. We hypothesize that past trajectories contain enough signal to guide skill learning, even without replay or validation, provided the agent can learn to analyze them.

智能体可以在推理阶段通过增量式地发现这些程序，并将其装备到未来的任务中进行学习。我们假设，即使没有重放或验证，只要智能体能够学会分析过去的轨迹，这些轨迹中就包含了足够的信号来指导技能学习。

To test our claims, we propose SpeedRunner, a coding agent that analyzes trajectories and refactors skills for better performance on future tasks. Across three different embodied environments, we show that SpeedRunner consistently achieves the frontier in learning and cost reduction while remaining robust against distribution shifts and environmental randomness.

为了验证我们的观点，我们提出了 SpeedRunner，这是一个能够分析轨迹并重构技能以在未来任务中获得更好表现的编码智能体。在三个不同的具身智能环境中，我们证明了 SpeedRunner 在学习效率和成本削减方面始终处于领先地位，同时对分布偏移和环境随机性保持稳健。