AI RESEARCH
LaMP: Learning Vision-Language-Action Policies with 3D Scene Flow as Latent Motion Prior
arXiv CS.CV
•
ArXi:2603.25399v1 Announce Type: new