Fetching the paper…

UniSkill: Imitating Human Videos via Cross-Embodiment Skill Representations · Around