Preprint
Aug 2026
DECOWAM: Decoupled Whole-Body World-Action Model for Legged Mobile Manipulation
DECOWAM is introduced, a whole-body world-action model that separates camera ego-motion from base and arm actions through dedicated conditional interfaces and shows that embodiment-aware factorization can support parameter-efficient joint visual prediction and whole-body control under moving viewpoints.
Siyuan Ma, Boshi Zhang, Yutian Zhang et al.
· 1 citation