论文/arXiv 2610.17526·15 天前

Box-WAM：边界框轨迹作为世界动作模型的紧凑状态表示

BoxWAM用任务相关物体的有向边界框轨迹替代未来视频帧，在机器人操作任务上降低计算成本并提升动作预测。

AI 整理

• 本体 SDK 与控制
• 端侧部署

原文链接：Box-WAM: Box Trajectories as Compact State Representations for World Action Models（arXiv 2610.17526 · 发布日期：2026-10-13）

Box-WAM: Box Trajectories as Compact State Representations for World Action Models

BoxWAM用任务相关物体的有向边界框轨迹替代未来视频帧，在机器人操作任务上降低计算成本并提升动作预测。

核心要点

• WAM显式预测动作和未来状态，但全帧视频预测在训练和推理中计算开销较高。
• BoxWAM以任务相关物体的有向边界框轨迹表示未来，捕捉动作引起的关键物理变化。
• 结果显示，框轨迹预测显著降低计算成本，同时改善动作预测性能。
• 摘要提供结果视频链接，未说明代码或模型是否开源。

验证范围与局限：仅说明在机器人操作任务上验证，摘要未注明仿真或真机范围。

英文摘要 (Abstract)

World action models (WAMs) are a category of policy models that ex- plicitly predict both actions and future world states. They have been shown to outperform action-only models across various robotic manipulation tasks. How- ever, representing the future with full-frame video prediction can be computation- ally expensive during both training and inference. This raises a central question: what compact state representation preserves the planning and supervision bene- fits of future prediction? In this paper, we propose BoxWAM, which replaces dense future frames with oriented bounding box trajectories of task-relevant ob- jects. By explicitly imagining object-centric future motion, our method provides a lightweight prediction target that captures the key physical changes induced by robot actions. We show that box trajectory prediction substantially reduces com- putational cost while improving action prediction performance, suggesting that object-centric compact states can serve as an efficient alternative to video-based world action modeling for robotic manipulation. Result videos can be found on https://box-wam.github.io

作者：Qinxi Yu, 丁文舟, Gloria X. Wang, Albert J. Zhai, Shenlong Wang

来源

本页事实对应的公开页面，以原文为准。

1. arXiv 2610.17526：Box-WAM: Box Trajectories as Compact State Representations for World Action Models访问 2026-10-21