用推理引导想象未来，让机器人更准地规划长序列动作。
RoboTALES: Learning Reasoning-Guided Robot Policies via Task-Aligned Simulated Futures
- 用分层大模型分解任务目标，引导视频生成聚焦任务意图
- 通过视觉语言模型评判想象轨迹，用奖励反馈保持目标一致性
- 在复杂操作任务中表现优于现有方法，尤其适合长程规划
预训练视频生成模型是视觉-运动控制的有力候选，但其生成的未来常常偏离任务目标，且难以可靠地受动作条件约束。为此，我们提出RoboTALES，一种单阶段框架，通过学习与任务对齐的模拟未来来训练机器人策略。该方法引入两项关键创新：（1）基于分层大语言模型的规划器，将复杂任务拆解为一系列子目标以引导模型的想象；（2）基于视觉语言模型的评判器，评估这些“想象”轨迹，并利用奖励反馈使模型内部表征持续聚焦于目标。通过将视频生成锚定在抽象推理上，我们实现了时间上一致的动作推演和更连贯的行为。我们在RoboCasa和LIBERO10的多种操作任务上评估了RoboTALES，结果表明该方法在长时程任务中持续优于现有方法。代码与模型已公开于https://github.com/hananshafi/RoboTALES。
原文摘要 · Abstract (English)
Pretrained video generative models are promising backbones for visuomotor control, but their imagined futures often drift from task intent and are not reliably action-conditional. As a result, these models can be difficult to use for planning or policy extraction. To address these limitations, we propose RoboTALES, a single-stage framework that learns task-aligned simulated futures and uses them to train robot policies. Our approach introduces two key innovations: (1) a hierarchical LLM-based planner that breaks complex tasks into a sequence of subgoals to guide the model's imagination; and (2) a VLM-based critic that evaluates these ``imagined'' futures and uses reward-based feedback to keep the model's internal representations focused on the goal. By anchoring the video generator in abstract reasoning, we produce temporally consistent rollouts and more coherent actions. We evaluate RoboTALES on diverse manipulation tasks from RoboCasa and LIBERO10, and show that our method consistently outperforms existing methods, especially in long-horizon tasks. Our code and models are publicly available at https://github.com/hananshafi/RoboTALES.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。