Paper Detail
Weijie Wang, Xiaoxuan He, Youping Gu, Yifan Yang, Zeyu Zhang, Yefei He, Yanbo Ding, Xirui Hu, Donny Y. Chen, Zhiyuan He, Yuqing Yang, Bohan Zhuang
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.
No structured notes yet. Add `summary_sections`, `why_relevant`, `claim_impact`, or `next_action` in `papers.jsonl` to enrich this view.
No ranking explanation is available yet.
No tags.
@article{wang2026world,
title = {World-R1: Reinforcing 3D Constraints for Text-to-Video Generation},
author = {Weijie Wang and Xiaoxuan He and Youping Gu and Yifan Yang and Zeyu Zhang and Yefei He and Yanbo Ding and Xirui Hu and Donny Y. Chen and Zhiyuan He and Yuqing Yang and Bohan Zhuang},
year = {2026},
abstract = {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. Utiliz},
url = {https://arxiv.org/abs/2604.24764},
keywords = {cs.CV, video foundation models, 3D priors, reinforcement learning, Flow-GRPO, pre-trained 3D foundation models, vision-language models, periodic decoupled training, geometric consistency, world simulation, code available, huggingface daily},
eprint = {2604.24764},
archiveprefix = {arXiv},
}
{}