cs.CVNov 12, 2026

WorldDraft: If Code Builds the World, Can Video Models Just Render It?

Authors: Zhiyu Hou, Wen-Jie Shu, Harold Haodong Chen, Weilin Ruan, Chenfei Liao, Liuzhou Zhang, Disen Lan, Xiaogang Xu, 

Organizations: SUSTech · ZODA · HKUST · CUHK · 

Abstract

Multimodal coding agents decouple 4D world construction from visual generation, repurposing video models from world generators toward generative renderers. This division of labor raises a fundamental question: if code builds the world, can video models just render it? A reliable generative renderer hinges on two complementary criteria: (i) world preservation, retaining prescribed 4D kinematics and camera paths; and (ii) visual realization, persistently binding text-specified target attributes over time. To evaluate this capability, we introduce WorldDraft, a benchmark of 432432 strictly paired test cases across 4848 executable 4D world families spanning six dynamic challenge categories. Under a 3×33 \times 3 factorial design, WorldDraft orthogonally crosses draft granularity (abstract proxy to previs) and appearance alignment (neutral to conflicting) under identical dynamics. Extensive evaluations across 1212 representative models reveal that: (I) world preservation and visual realization remain in stark operational tension; (II) richer draft scaffolding anchors structure but paradoxically impairs appearance fidelity; and (III) conflicting source cues trigger severe source leakage (up to 64%64\%), exposing fragile cross-modal disentanglement. To mitigate these bottlenecks, we propose WorldDraft-Primer, a plug-and-play visual prompting strategy that consistently improves both objectives across frozen video renderers. Together, WorldDraft and WorldDraft-Primer lay a foundation for bridging programmatic 4D simulation with video generation, advancing physically grounded generative rendering.

Figures & tables

Appendix figures & tables10 assets

Supplementary material from the paper’s appendix.

Appendix

Explore similar work

CardsList
  1. AEON: Unifying Video and 3D World Models

    Oct 20, 2026Weiqi Zhang, Wenyuan Zhang, Yu-Shen Liu +1Video World Models3D-Aware Video Generation

  2. Code Plans, Diffusion Renders: Open-Ended Generative World Modeling

    Sep 22, 2026Zixun Fang, Yawen Shao, Kai Zhu +7World ModelsCode World Models

  3. DreamWorld: Unified World Modeling in Video Generation

    Feb 28, 2026Boming Tan, Xiangdong Zhang, Ning Liao +73D-Aware Video GenerationVideo World Models