LIVE · refreshes every 20 min

updated Sep 16, 04:23 UTC

110010.art / archive / research / 2025-07

Research — July 2025

Jul 1 Predicting ego-centric video from human actions (PEVA) generates the next frame from past frames and actions; model supports atomic actions, counterfactuals, and long video generation BAIR (Berkeley) source ↗