ResearcharXivNEW

Hydra-0: Action Flow for Generalist World Modeling and Control

Li 2026-08-18
Hongyu LiBowen WenXinghao Zhu

We introduce Hydra-0, a generalist world model conditioned on action flow, which represents robot actions as pixel motion. This shared visual interface enables generalist world modeling and control by learning action consequences across embodiments, tasks, environments, and video-generation backbones. Our best configuration achieves 90.4% lower robot-motion error and 60.2% lower object-motion erro

Read on arXiv
Data aggregated and editorially reviewed by TrendMing.

Key Contributions

  • We introduce Hydra-0, a generalist world model conditioned on action flow, which represents robot actions as pixel motion.
  • This shared visual interface enables generalist world modeling and control by learning action consequences across embodiments, tasks, environments, and video-generation backbones.
  • Our best configuration achieves 90.4% lower robot-motion error and 60.2% lower object-motion erro

Research Themes

AIResearch