Cosmos Policy: 为视觉运动控制和规划微调视频模型
Cosmos Policy: Fine-Tuning Video Models for Visuomotor Control and Planning
AI总结：
Cosmos Policy通过单阶段后训练将预训练视频模型转化为高效机器人策略，实现视觉运动控制与规划的先进性能。
摘要：
最近的视频生成模型展示了在捕捉复杂物理交互和场景随时间演变方面的非凡能力。为了利用其时空先验知识，机器人研究工作将视频模型适应于策略学习，但引入了复杂性，因为需要多个阶段的后训练和新的架构组件来生成动作。在本工作中，我们介绍了Cosmos Policy，这是一种简单的方法，通过在目标平台上收集的机器人示范数据上进行一次后训练，将预训练的大型视频模型（Cosmos-Predict2）适应为有效的机器人策略，而无需进行架构修改。Cosmos Policy学习直接生成编码为潜在帧的机器人动作，这些动作在视频模型的潜在扩散过程中。利用模型的预训练先验知识和核心学习算法，捕捉复杂的动作分布。此外，Cosmos Policy生成未来状态图像和价值（预期累积奖励），这些同样编码为潜在帧，使在测试时能够通过更高的成功率规划动作轨迹。在我们的评估中，Cosmos Policy在LIBERO和RoboCasa模拟基准（分别达到98.5%和67.1%的平均成功率）以及具有挑战性的现实世界双臂操作任务中取得了最先进的性能，优于从头训练的强扩散策略、基于视频模型的策略以及在相同机器人示范上微调的最先进的视觉-语言-动作模型。此外，给定策略滚动数据，Cosmos Policy可以学习经验来改进其世界模型和价值函数，并利用基于模型的规划在具有挑战性的任务中实现更高的成功率。我们发布了代码、模型和训练数据，网址为https://research.nvidia.com/labs/dir/cosmos-policy/
英文摘要：
Recent video generation models demonstrate remarkable ability to capture complex physical interactions and scene evolution over time. To leverage their spatiotemporal priors, robotics works have adapted video models for policy learning but introduce complexity by requiring multiple stages of post-training and new architectural components for action generation. In this work, we introduce Cosmos Policy, a simple approach for adapting a large pretrained video model (Cosmos-Predict2) into an effective robot policy through a single stage of post-training on the robot demonstration data collected on the target platform, with no architectural modifications. Cosmos Policy learns to directly generate robot actions encoded as latent frames within the video model's latent diffusion process, harnessing the model's pretrained priors and core learning algorithm to capture complex action distributions. Additionally, Cosmos Policy generates future state images and values (expected cumulative rewards), which are similarly encoded as latent frames, enabling test-time planning of action trajectories with higher likelihood of success. In our evaluations, Cosmos Policy achieves state-of-the-art performance on the LIBERO and RoboCasa simulation benchmarks (98.5% and 67.1% average success rates, respectively) and the highest average score in challenging real-world bimanual manipulation tasks, outperforming strong diffusion policies trained from scratch, video model-based policies, and state-of-the-art vision-language-action models fine-tuned on the same robot demonstrations. Furthermore, given policy rollout data, Cosmos Policy can learn from experience to refine its world model and value function and leverage model-based planning to achieve even higher success rates in challenging tasks. We release code, models, and training data at https://research.nvidia.com/labs/dir/cosmos-policy/
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
