V-JEPA 2: Self-Supervised Video Models Enable Understanding,Prediction,Planninggithub.com6 points·johlo··0 commentsOpen articleSaveView on HN