World-R1: Reinforcing 3D Constraints for Text-to-Video Generation
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866914604218580992 |
|---|---|
| author | Wang, Weijie He, Xiaoxuan Gu, Youping Yang, Yifan Zhang, Zeyu He, Yefei Ding, Yanbo Hu, Xirui Chen, Donny Y. He, Zhiyuan Yang, Yuqing Zhuang, Bohan |
| author_facet | Wang, Weijie He, Xiaoxuan Gu, Youping Yang, Yifan Zhang, Zeyu He, Yefei Ding, Yanbo Hu, Xirui Chen, Donny Y. He, Zhiyuan Yang, Yuqing Zhuang, Bohan |
| contents | Recent video foundation models demonstrate impressive visual synthesis but frequently suffer from geometric inconsistencies. While existing methods attempt to inject 3D priors via architectural modifications, they often incur high computational costs and limit scalability. We propose World-R1, a framework that aligns video generation with 3D constraints through reinforcement learning. To facilitate this alignment, we introduce a specialized pure text dataset tailored for world simulation. Utilizing Flow-GRPO, we optimize the model using feedback from pre-trained 3D foundation models and vision-language models to enforce structural coherence without altering the underlying architecture. We further employ a periodic decoupled training strategy to balance rigid geometric consistency with dynamic scene fluidity. Extensive evaluations reveal that our approach significantly enhances 3D consistency while preserving the original visual quality of the foundation model, effectively bridging the gap between video generation and scalable world simulation. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2604_24764 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | World-R1: Reinforcing 3D Constraints for Text-to-Video Generation Wang, Weijie He, Xiaoxuan Gu, Youping Yang, Yifan Zhang, Zeyu He, Yefei Ding, Yanbo Hu, Xirui Chen, Donny Y. He, Zhiyuan Yang, Yuqing Zhuang, Bohan Computer Vision and Pattern Recognition Recent video foundation models demonstrate impressive visual synthesis but frequently suffer from geometric inconsistencies. While existing methods attempt to inject 3D priors via architectural modifications, they often incur high computational costs and limit scalability. We propose World-R1, a framework that aligns video generation with 3D constraints through reinforcement learning. To facilitate this alignment, we introduce a specialized pure text dataset tailored for world simulation. Utilizing Flow-GRPO, we optimize the model using feedback from pre-trained 3D foundation models and vision-language models to enforce structural coherence without altering the underlying architecture. We further employ a periodic decoupled training strategy to balance rigid geometric consistency with dynamic scene fluidity. Extensive evaluations reveal that our approach significantly enhances 3D consistency while preserving the original visual quality of the foundation model, effectively bridging the gap between video generation and scalable world simulation. |
| title | World-R1: Reinforcing 3D Constraints for Text-to-Video Generation |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2604.24764 |