DobotTrack: Target-Grounded Vision-Language-Action Learning for Embodied Person Tracking Paper • 2611.09041 • Published 10 days ago • 1