机器人学动作不再需要人教，看视频就行
机器人学动作一直卡在一个死结：要学得准，就得靠人亲手标注的示范数据，但这类数据又少又贵；网上人类视频多到看不完，可里面没有“动作标签”，机器人不知道手该往哪放。这篇把“动作”重新定义成一条“动态链”——从当前状态到未来状态中间的一串离散变化，而不是某个具体的机械臂坐标。这样一来，人类视频里的画面变化本身就变成了监督信号：机器人先猜一条动态链，再想象它会导致什么结果，如果想象出来的画面和视频里真实发生的对上了，这条链就被强化。等于把“看人做事”和“自己动手”塞进了同一个推理空间，机器人能从海量人类视频里学，再迁移到自己的机械臂上。实验里，它在没见过的任务和环境上表现明显更好。这不是你明天能用的东西，但它指向一个方向：机器人不再需要人类手把手教，看就够了。
📄 原文摘要(英文)
Vision-Language-Action (VLA) models have made substantial progress in robotic manipulation, but their generalization remains constrained by the diversity and coverage of action-labeled robot demonstrations. Human videos provide much broader behavioral and environmental experience, while learning from them faces a fundamental trade-off between scalable data and explicit action grounding. We introduce ThinkWorld, a VLA learning framework that grounds action reasoning in world dynamics and strengthens it through reinforcement learning with world imagination. ThinkWorld represents each action thought as a Chain-of-Dynamics (CoD), a sequence of discrete motion tokens that captures the intermediate dynamics connecting the current world to a future state and serves as a common action representation across embodiments. We first initialize CoD with action-labeled demonstrations, then predict the future induced by each CoD and reinforce those whose predictions better match the observed outcome. This enables videos without action annotations to provide direct supervision from observed state transitions, allowing human and robot experience to refine the same reasoning space. Experiments in simulation and on real robots demonstrate effective human-to-robot transfer and strong generalization to unseen tasks and environments.