EgoWAM：用野外第一视角人类数据训练超越像素的世界动作模型

View PDF HTML (experimental)

Abstract:Egocentric human data offers scalable supervision for robot manipulation. However, behavior cloning entangles transferable content like objects, scenes, and task semantics, with non-transferable factors like human morphology, head motion, and behavioral style. We study whether World Action Models (WAMs) provide a better training signal by requiring policies to predict not only actions, but also how the scene evolves. The central question is what world representation best enables human-to-robot transfer. We hypothesize that an effective world target should abstract appearance, capture agent-invariant physical effects, and separate camera motion from environment change. We introduce EgoWAM, a controlled human-robot co-training framework that fixes the policy backbone, action head, and data mixture while varying only the world prediction target, comparing Pixel, DINO, and 3D motion flow. Across three real-world bimanual tasks, WAM co-training scales more effectively with in-the-wild egocentric human data than behavior cloning. Pixel-based prediction transfers weakly, while DINO and 3D flow yield substantial gains: DINO improves out-of-distribution object and scene generalization by up to 4x, and 3D flow improves in-domain performance by 20-30%. More details: this https URL

Subjects:	Robotics (cs.RO); Artificial Intelligence (cs.AI)
Cite as:	arXiv:2607.08436 [cs.RO]
	(or arXiv:2607.08436v2 [cs.RO] for this version)
	https://doi.org/10.48550/arXiv.2607.08436 arXiv-issued DOI via DataCite

Submission history

From: Baoyu Li [view email]
[v1] Wed, 8 Jul 2026 16:11:37 UTC (82,547 KB)
[v2] Tue, 13 Oct 2026 05:21:40 UTC (27,189 KB)