works on
cross-embodiment 1multimodal learning 1real-time inference 1robot control 1video-action prediction 1world modeling 1

From the 1 of 1 linked paper with an AI index.

collaborators

1 paper

cs.RO2026

Motubrain: An Advanced World Action Model for Robot Control

Motubrain Team, Chendong Xiang, Fan Bao +17

Motubrain is a unified world action model that jointly learns video and robot actions using a UniDiffuser and Mixture-of-Transformers architecture, enabling policy learning, world…