Hugging Face Daily Papers·· 1 day agoAI score38
Coda-DaVinci: Rethinking Video Generation with Code2Video
Coda-DaVinci: Reimagining Video Generation with Code2Video
AI briefing
Coda-DaVinci proposes a task-adaptive Code2Video agent framework that can choose between agent-guided and direct generation and adjusts modeling granularity according to entities. Experiments show that it leads in VTR instruction following and R2M scene consistency, has better physical plausibility than direct generation, and higher visual realism than comparison baselines. In long-video evaluation, event structure and cross-interaction consistency improve, but transition stability and scene-setting adherence remain limited.
Open original arxiv.org