Paper 2604.24764
World-R1: Reinforcing 3D Constraints for Text-to-Video Generation
- Published
- Apr 2026
- Research lab
- Independent
- Citations
- 7
- GitHub
- 414 stars
01 In brief
Summary
World-R1 is a framework that improves 3D consistency in text-to-video generation by using reinforcement learning (RL) instead of architectural changes.
It builds on the Wan 2.1 model and uses Flow-GRPO to optimize the model with rewards from 3D foundation models (Depth Anything 3) and vision-language models (Qwen3-VL).
The reward system includes meta-view scoring, reconstruction fidelity, and trajectory alignment, plus a general aesthetic reward.
Camera control is achieved via implicit noise wrapping, and a pure text dataset of about 3,000 prompts is used for training.
A periodic decoupled training strategy balances rigid geometry with dynamic scene fluidity.
Experiments show World-R1 significantly improves 3D consistency (PSNR gains of 10.23dB and 7.91dB over baselines) while maintaining high visual quality on VBench.
A user study shows 92% preference for geometric consistency and 86% overall preference over the base model.
The method avoids expensive 3D datasets and inference-time constraints, making it scalable for world simulation applications like autonomous driving.
02 From the paper
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. 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.