hao05/coconut

Dr. Seg: Revisiting GRPO Training for Visual Large Language Models through Perception-Oriented Design

This repository contains the weights for Dr. Seg-7B, as presented in the paper Dr. Seg: Revisiting GRPO Training for Visual Large Language Models through Perception-Oriented Design.

Dr. Seg is a plug-and-play GRPO-based framework designed to adapt Visual Large Language Models (VLLMs) for visual perception tasks such as reasoning segmentation and object detection. It introduces two key components: a Look-to-Confirm mechanism and a Distribution-Ranked Reward module, requiring no architectural modifications and integrating seamlessly with existing GRPO-based VLLMs.

Links

• Paper: arXiv:2603.00152
• Dataset: COCONut
• Code: GitHub Repository

Model Description

Dr. Seg-7B is fine-tuned from Qwen2.5-VL-7B-Instruct using perception-oriented designs. While standard GRPO is often tailored for language reasoning, Dr. Seg addresses the specific needs of visual perception by providing a broader output space and fine-grained, stable reward signals. Experiments demonstrate that Dr. Seg improves performance in complex visual scenarios while maintaining strong generalization.

Citation

If you find this work useful, please cite:

@article{sun2026dr,
  title={Dr. Seg: Revisiting GRPO Training for Visual Large Language Models through Perception-Oriented Design},
  author={Sun, Haoxiang and Wang, Tao and Tang, Chenwei and Yuan, Li and Lv, Jiancheng},
  journal={arXiv preprint arXiv:2603.00152},
  year={2026}
}

Acknowledgements

This project builds upon several open-source efforts, including VisionReasoner, Seg-Zero, EasyR1, veRL, and COCONut-PanCap. We also utilize pretrained models from Qwen2.5-VL and SAM2.