Institution

ACE Robotics

An embodied-AI and robotics lab working on vision-language-action models, behind the ACE-Ego-0 VLA pretraining framework.

Vision-Language-Action · ACE Robotics

ACE-Ego-0: Unifying Egocentric Human and Robot Data for VLA Pretraining

ACE-Ego-0 pretrains a VLA on 6,000+ hours mixing robot trajectories with human egocentric video turned into pseudo-actions. It averages 78.3% on six real bimanual tasks vs 71.7% for pi-0.5 and 35.6% for GR00T-N1.7.