机器人学技能，靠的是给每个技能写清“依赖什么”
机器人学新技能通常只靠几个示范，但真正难的是两件事：把学过的技能重新组合去干新活，以及让每个技能在没见过的场景里也能用。这两件事互相依赖，可中间信息断了——组合层只看到技能的名字或指令，看不到这个技能到底在什么条件下成立。这篇论文的做法很直接：把每个技能“依赖什么”写清楚，比如“抓取只取决于手爪相对物体的姿态”，训练时就把这个结构嵌进去，组合时也拿它当接口。系统会为每个技能生成多个候选结构、分别训练验证，运行时再按任务目标挑合适的拼起来。在 MetaWorld 和长程 ManiSkill 任务上，它让技能在没见过的情况下也能用，还能拼出从未见过的组合；去掉这层接口信息，性能明显掉。它不是你明天能用上的东西，但给了一个值得记住的思路：与其让 AI 默默学会，不如逼它说出自己靠什么。
📄 原文摘要(英文)
Robots that learn from a few demonstrations often require two forms of generalization. Compositional generalization recombines skills to solve new tasks, and skill generalization lets the learned policy behind each skill work in new situations. The two depend on each other, yet information is lost between composition and the skills it calls. Where a skill works is determined by the structure its policy is trained with, while composition sees the skill only through a separate description, such as a name, an instruction, or a symbolic operator, that omits this structure. Our key idea is to use each policy's structural prior as part of the interface between composition and the skill. A structural prior states what a behavior depends on, for example that a grasp depends only on the gripper's pose relative to the object. Built into training, it shapes where the policy generalizes; stated in language, it tells composition where the policy applies. We instantiate this idea in Agent Priors-guided Policy Learning (APPL). A construction agent segments complete demonstrations into reusable skills, proposes several structural priors for each skill, and trains and verifies one policy per prior. A runtime agent then selects among these prior-specific policies and composes them toward new task goals using their interfaces. Across MetaWorld and long-horizon ManiSkill tasks, APPL improves out-of-distribution skill generalization and enables previously unseen skill compositions; ablating the interface information substantially reduces performance. These results support the use of training-time structural assumptions as a bridge between skill learning and skill composition.