metadata

base_model:
  - Qwen/Qwen2.5-VL-7B-Instruct
language:
  - en
license: apache-2.0
tags:
  - transformers
  - multimodal
pipeline_tag: visual-question-answering

VL-Reasoner-7B

🚀 News: We release our meticulously curated collection of RL training queries for multimodal reasoning: ViRL39K.

VL-Reasoner-7B achieves superior results on various multimodal reasoning benchmarks.

It is trained using the GRPO-SSR techniques, serving as the foundation for VL-Rethinker.

For details of our approach and performance comparison, please see our paper.

For details of training and evaluation, please see our code repo.

Explore further via the following links:

Citation

If you feel this model useful, please give us a free cite:

@article{vl-rethinker,
      title={VL-Rethinker: Incentivizing Self-Reflection of Vision-Language Models with Reinforcement Learning},
      author = {Wang, Haozhe and Qu, Chao and Huang, Zuming and Chu, Wei and Lin, Fangzhen and Chen, Wenhu},
      journal={arXiv preprint arXiv:2504.08837},
      year={2025}
}