AI RESEARCH
R3D: Revisiting 3D Policy Learning
arXiv CS.CV
•
ArXi:2604.15281v1 Announce Type: new 3D policy learning promises superior generalization and cross-embodiment transfer, but progress has been hindered by