DreamX: Video Diffusion Model is a Universal Latent Action Learner

Learning a universal action representation from video requires modeling dynamics that are consistent across different executions of the same action ye…

Continue in Briflio to read, react, comment, and share.