发表机构
Fudan University; Shanghai Innovation Institute; The Chinese University of Hong Kong; NovaXBot（复旦大学; 上海创新研究院; 香港中文大学; NovaXBot）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
Vela提出基于样条的连续动作表示，通过自适应时间分辨率解决固定输出预算的局限，在仿真和真实任务中验证了其作为具身基础模型基础的潜力。
AI 中文摘要
大多数视觉-语言-动作模型将未来运动表示为固定速率的动作块，将时间分辨率和预测范围绑定到固定的输出预算。这种逐点表示在高度相关的相邻动作上浪费容量，将时间连续性和平滑性留给隐式学习，并迫使在长时程覆盖与接触丰富操作所需的局部精度之间进行权衡。为解决这些局限性，我们引入了Vela，一个将未来机器人行为表示为连续轨迹的视觉-语言-动作基础模型。Vela结合了基于紧凑样条的动作表示、运动相关的时间支持以及用于异构本体的共享动作接口，使得固定的输出预算能够跨运动调整其时间分辨率。我们在大规模多本体机器人数据上预训练Vela，并在LIBERO-X、EBench以及两个真实世界长时程任务（蛋饼烹饪和土豆切丝）上进行了评估，在仿真和物理操作中取得了有前景的结果。这些结果凸显了连续动作表示作为未来具身基础模型基础的潜力。项目页面和更多结果：此HTTPS URL。
英文摘要
Most vision-language-action models represent future motion as fixed-rate action chunks, tying temporal resolution and prediction horizon to a fixed output budget. This pointwise representation wastes capacity on highly correlated neighboring actions, leaves temporal continuity and smoothness to be learned implicitly, and forces a tradeoff between long-horizon coverage and the local precision required for contact-rich manipulation. To address these limitations, we introduce Vela, a vision-language-action foundation model that represents future robot behavior as continuous trajectories. Vela combines a compact spline-based action representation with motion-dependent temporal support and a shared action interface for heterogeneous embodiments, allowing a fixed output budget to adapt its temporal resolution across motions. We pretrain Vela on large-scale multi-embodiment robot data and evaluate it on LIBERO-X, EBench, and two real-world long-horizon tasks, egg-cake cooking and potato shredding, obtaining promising results across simulation and physical manipulation. These results highlight the potential of continuous action representations as a foundation for future embodied foundation models. Project page and more results: https://clementine24.github.io/Vela/ .
