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