@SandeepRoutra11
๐ Excited to share ViPRA: Video Prediction for Robot Actions ๐ Accepted to #ICLR2026 @iclr_conf ๐ Best Paper โ #NeurIPS2025 Embodied World Models Workshop Robot learning today still needs millions of action labeled videos. Yet videos are abundant โ from humans and the web โ but lack action labels. Meanwhile, pretrained video models already learn rich dynamics. ViPRA is a recipe for turning pretrained video models into robot policies while enabling robot learning to scale with actionless videos. ๐งต Thread โ