PointVLA: Injecting the 3D World into Vision-Language-Action Models

Open in new window