| --- |
| license: apache-2.0 |
| library_name: diffusers |
| pipeline_tag: image-to-video |
| base_model: |
| - Wan-AI/Wan2.2-TI2V-5B |
| datasets: |
| - DarthZhu/VideoRLVR-Data |
| --- |
| |
| # VideoRLVR |
|
|
| VideoRLVR is a reinforcement learning (RL) recipe for training video reasoning models with verifiable rewards, introduced in the paper [Video Models Can Reason with Verifiable Rewards](https://huggingface.co/papers/2605.15458). |
|
|
| This checkpoint is an RL-optimized version of [Wan2.2-TI2V-5B](https://huggingface.co/Wan-AI/Wan2.2-TI2V-5B) trained on procedurally generated reasoning tasks including Maze, FlowFree, and Sokoban. |
|
|
| - **Paper:** [Video Models Can Reason with Verifiable Rewards](https://huggingface.co/papers/2605.15458) |
| - **Project Page:** [https://darthzhu.github.io/VideoRLVR-page/](https://darthzhu.github.io/VideoRLVR-page/) |
| - **Repository:** [https://github.com/luka-group/VideoRLVR](https://github.com/luka-group/VideoRLVR) |
|
|
| ## Overview |
|
|
| VideoRLVR formulates video reasoning as the generation of verifiable visual trajectories. It utilizes an SDE-GRPO optimization backbone, dense decomposed rewards, and an Early-Step Focus strategy for efficient training. This approach enables video diffusion models to satisfy explicit spatial, temporal, or logical constraints, moving beyond perceptual imitation toward reliable rule-consistent visual reasoning. |
|
|
| Across tasks like Maze, FlowFree, and Sokoban, VideoRLVR consistently improves over supervised fine-tuning baselines, demonstrating that verifiable RL can effectively optimize models for objective success criteria. |
|
|
| ## Citation |
|
|
| ```bibtex |
| @article{zhu2026video, |
| title={Video Models Can Reason with Verifiable Rewards}, |
| author={Tinghui Zhu and Sheng Zhang and James Y. Huang and Selena Song and Xiaofei Wen and Yuankai Li and Hoifung Poon and Muhao Chen}, |
| journal={arXiv preprint arXiv:2605.15458}, |
| year={2026} |
| } |
| ``` |
