1 paper · 1 filter
Yifan Li, Yukai Gu, Yingqian Min +6
Recent breakthroughs in video generation have demonstrated an emerging capability termed Chain-of-Frames (CoF) reasoning, where models resolve complex tasks through the generation…