VLA-JEPA: Enhancing Vision-Language-Action Model with Latent World Model Paper • 2602.10098 • Published about 1 month ago • 19