CoF-T2I: Video Models as Pure Visual Reasoners for Text-to-Image Generation Paper • 2601.10061 • Published Jan 15 • 31
