ViCrit: A Verifiable Reinforcement Learning Proxy Task for Visual Perception in VLMs

🤗ViCrit-Bench • 📊ViCrit Training Dataset• 📄Paper

Official codebase of "ViCrit: A Verifiable Reinforcement Learning Proxy Task for Visual Perception in VLMs". We will release all codes in one week.

If you found this work useful, consider giving this repository a star and citing our paper as followed:

@article{wang2025vicrit,
  title={ViCrit: A Verifiable Reinforcement Learning Proxy Task for Visual Perception in VLMs},
  author={Wang, Xiyao and Yang, Zhengyuan and Feng, Chao and Liang, Yongyuan and Zhou, Yuhang and Liu, Xiaoyu and Zang, Ziyi and Li, Ming and Lin, Chung-Ching and Lin, Kevin and others},
  journal={arXiv preprint arXiv:2506.10128},
  year={2025}
}