---
title: "VLM-R1 — paper by Shen et al. (2025) — summary and ideas"
description: "The authors build VLM-R1, an open framework on top of open-r1. 13 references, cited by 66 papers here."
authors: ["Haozhan Shen", "Peng Liu", "Jingcheng Li", "Chunxin Fang", "Yibo Ma", "Jiajia Liao", "Qiaoli Shen", "Zilun Zhang", "Kangjia Zhao", "Qianqian Zhang", "Ruochen Xu", "Tiancheng Zhao"]
year: 2025
venue: "arXiv:2504.07615"
source: "https://arxiv.org/abs/2504.07615"
doi: "10.48550/arxiv.2504.07615"
arxiv: "2504.07615"
licence: "CC BY-SA 4.0 (arXiv)"
licence_url: "http://creativecommons.org/licenses/by-sa/4.0/"
summary_licence: "Elamur Cosmic License"
canonical: "https://research.elamur.ai/free/en/papers/shen-2025-vlm-r1-stable-generalizable-r1/"
---
[Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/shen-2025-vlm-r1-stable-generalizable-r1/)

[home](https://md.research.elamur.ai/free/en/) / [papers](https://md.research.elamur.ai/free/en/papers/) / [2025](https://md.research.elamur.ai/free/en/papers/2025/page/2/) / VLM-R1: A Stable and Generalizable R1-style Large Vision-Language Model

# VLM-R1: A Stable and Generalizable R1-style Large Vision-Language Model

**Authors:** Haozhan Shen, [Peng Liu](https://md.research.elamur.ai/free/en/authors/peng-liu/), Jingcheng Li, Chunxin Fang, Yibo Ma, Jiajia Liao, Qiaoli Shen, Zilun Zhang, Kangjia Zhao, Qianqian Zhang, [Ruochen Xu](https://md.research.elamur.ai/free/en/authors/ruochen-xu/) and [Tiancheng Zhao](https://md.research.elamur.ai/free/en/authors/tiancheng-zhao/) · **Year:** 2025 · **Venue:** arXiv:2504.07615 · **Licence:** [CC BY-SA 4.0 (arXiv)](http://creativecommons.org/licenses/by-sa/4.0/)

**Links:** [Publisher's page](https://arxiv.org/abs/2504.07615) · [PDF](https://arxiv.org/pdf/2504.07615) · [DOI 10.48550/arxiv.2504.07615](https://doi.org/10.48550/arxiv.2504.07615) · [arXiv 2504.07615](https://arxiv.org/abs/2504.07615) · [OpenAlex](https://openalex.org/W6948439803)

**Citations:** 4 citations (OpenAlex, 2026-10-14) · [66](https://md.research.elamur.ai/free/en/papers/shen-2025-vlm-r1-stable-generalizable-r1/cited-by/) in our library, of the 149,439 papers whose bibliographies we read

## Summary
The paper asks whether DeepSeek R1-style reinforcement learning with rule-based rewards can help vision-language models (VLMs) on perception tasks. Many such tasks have exact ground-truth labels, and VLMs often trail much smaller specialist detectors on them. The authors build VLM-R1, an open framework on top of open-r1. It supports GRPO, LoRA, multi-node and multi-image training, mixed image-text and text-only data, and the QwenVL and InternVL model families. They apply it to two tasks that both output bounding boxes: comprehending a referring expression to locate the described object (REC), and detecting objects drawn from an open vocabulary (OVD). The training data contains no reasoning annotations, and rewards are computed directly from box quality (e.g., IoU). Compared with SFT, the RL models do better and transfer especially well to hard out-of-domain benchmarks. The ablations show three things: a naive mAP reward invites reward hacking, data with richer semantics helps, and larger base models gain more from RL. These findings offer practical advice for reward and data design in multimodal RL.

## Key points
- Base models such as Qwen2.5-VL-3B (3.75B parameters) underperform Grounding DINO (341M), for example 37.5 vs 55.0 on ODinW. This gap motivates the RL approach.
- On REC, the 3B RL model reaches 63.16 on the out-of-domain LISA-Grounding benchmark, compared with 54.82 for SFT.
- On OVD, the 3B RL model gets 21.1 AP on COCO (17.8 for SFT, 14.2 for the 7B baseline). It also sets a new SOTA of 31.01 nms-AP on OVDEval (26.50 for SFT, 29.08 for the 7B model).
- **Reward hacking and the fix:** the plain mAP reward pushes the model to output many redundant boxes. The proposed odLength reward penalizes this behavior. With it, the model shows an "OD aha moment": it first reasons about whether the object is present and only then predicts boxes.
- **Data and scale:** training on D³ beats training on COCO, even on the in-domain COCO filtered set. Across model sizes, RL clearly improves the Relation sub-task (e.g., 20.6→27.0 for 32B). On Negation, however, the 3B model drops slightly (38.7→37.7). Code and weights are released at github.com/om-ai-lab/VLM-R1.

*Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.*

📚 Shelves (3): [🖥️ Computer Science](https://md.research.elamur.ai/free/en/shelves/computer-science/) · 18,614 works · [🤖 Artificial Intelligence](https://md.research.elamur.ai/free/en/shelves/computer-science/artificial-intelligence/) · 17,965 works · [👁️ Computer Vision and Pattern Recognition](https://md.research.elamur.ai/free/en/shelves/computer-science/computer-vision-and-pattern-recognition/) · 3,486 works

🧭 Topics (2): [🧠 Language models & pretraining](https://md.research.elamur.ai/free/en/topics/language-models/) · 3,894 works · [🖼️ Vision & language](https://md.research.elamur.ai/free/en/topics/multimodal/) · 811 works

## Cite

Cite as: Haozhan Shen, Peng Liu, Jingcheng Li, Chunxin Fang, Yibo Ma, Jiajia Liao, Qiaoli Shen, Zilun Zhang, Kangjia Zhao, Qianqian Zhang, Ruochen Xu and Tiancheng Zhao. 2025. VLM-R1: A Stable and Generalizable R1-style Large Vision-Language Model. arXiv:2504.07615.

The paper itself, not this page. BibTeX from the library's record of the work.

```bibtex
@misc{shen2025vlm,
  title = {{VLM-R1: A Stable and Generalizable R1-style Large Vision-Language Model}},
  author = {Haozhan Shen, Peng Liu, Jingcheng Li, Chunxin Fang, Yibo Ma, Jiajia Liao, Qiaoli Shen, Zilun Zhang, Kangjia Zhao, Qianqian Zhang, Ruochen Xu and Tiancheng Zhao},
  year = {2025},
  doi = {10.48550/arxiv.2504.07615},
  eprint = {2504.07615},
  archivePrefix = {arXiv},
  url = {https://doi.org/10.48550/arxiv.2504.07615}
}
```

- [← previous ToolSandbox: A Stateful, Conversational, Interactive Evaluation Benchmark for LLM Tool Use Capabilities](https://md.research.elamur.ai/free/en/papers/lu-2025-toolsandbox-stateful-conversational-interactive-evaluation/)
- [next → Group-in-Group Policy Optimization for LLM Agent Training](https://md.research.elamur.ai/free/en/papers/feng-2025-group-group-policy-optimization-llm/)
- [↑ back to papers of 2025](https://md.research.elamur.ai/free/en/papers/2025/page/2/)

## References

13 references, 8 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. Easyr1: An efficient, scalable, multi-modality rl training framework. [https://github.com/hiyouga/](https://github.com/hiyouga/) EasyR1, 2025. 4 [↗](https://github.com/hiyouga/)
2. Josh Achiam, Steven Adler, Sandhini Agarwal, Lama Ahmad, Ilge Akkaya, Florencia Leoni Aleman, Diogo Almeida, Janko Altenschmidt, Sam Altman, Shyamal Anadkat, et al. [Gpt-4 technical report](https://md.research.elamur.ai/free/en/papers/openai-2023-gpt-4-technical-report/). arXiv preprint arXiv:2303.08774, 2023. 3
3. Arash Ahmadian, Chris Cremer, Matthias Gallé, Marzieh Fadaee, Julia Kreutzer, Olivier Pietquin, Ahmet Üstün, and Sara Hooker. Back to basics: Revisiting reinforce style optimization for learning from human feedback in llms. arXiv preprint arXiv:2402.14740, 2024. 3 [find](https://scholar.google.com/scholar?q=Arash+Ahmadian%2C+Chris+Cremer%2C+Matthias+Galle%CC%81%2C+Marzieh+Fadaee%2C+Julia+Kreutzer%2C+Olivier+Pietquin%2C+Ahmet+U%CC%88stu%CC%88n%2C+and+Sara+Hooker%2E+Back+to+basics%3A+Revisiting+reinforce+style+optimization+for+learning+fr)
4. Jean-Baptiste Alayrac, Jeff Donahue, Pauline Luc, Antoine Miech, Iain Barr, Yana Hasson, Karel Lenc, Arthur Mensch, Katherine Millican, Malcolm Reynolds, et al. [Flamingo: a visual language model for few-shot learning](https://md.research.elamur.ai/free/en/papers/alayrac-2022-flamingo-visual-language-model-few/). Advances in neural information processing systems, 35:23716–23736, 2022. 3
5. Dario Amodei, Chris Olah, Jacob Steinhardt, Paul Christiano, John Schulman, and Dan Mané. [Concrete problems in ai safety](https://md.research.elamur.ai/free/en/papers/amodei-2016-concrete-problems-ai-safety/). arXiv preprint arXiv:1606.06565, 2016. 9
6. Shuai Bai, Keqin Chen, Xuejing Liu, Jialin Wang, Wenbin Ge, Sibo Song, Kai Dang, Peng Wang, Shijie Wang, Jun Tang, et al. Qwen2. 5-vl technical report. arXiv preprint arXiv:2502.[13923, 2025](https://md.research.elamur.ai/free/en/papers/bai-2025-qwen2-5-vl-technical/). 2, 3
7. Kezhen Chen, Rahul Thapa, Rahul Chalamala, Ben Athiwaratkun, Shuaiwen Leon Song, and James Zou. Dragonfly: Multi-resolution zoom supercharges large visual-language model. arXiv preprint arXiv:2406.00977, 2024. 3 — [arXiv:2406.00977](https://arxiv.org/abs/2406.00977) [find](https://scholar.google.com/scholar?q=Kezhen+Chen%2C+Rahul+Thapa%2C+Rahul+Chalamala%2C+Ben+Athiwaratkun%2C+Shuaiwen+Leon+Song%2C+and+James+Zou%2E+Dragonfly%3A+Multi%2Dresolution+zoom+supercharges+large+visual%2Dlanguage+model%2E+arXiv+preprint+arXiv%3A2406%2E009)
8. Lin Chen, Jisong Li, Xiaoyi Dong, Pan Zhang, Conghui He, Jiaqi Wang, Feng Zhao, and Dahua Lin. [Sharegpt4v: Improving large multi-modal models with better captions](https://md.research.elamur.ai/free/en/papers/chen-2023-sharegpt4v-improving-large-multi-modal/). arXiv preprint arXiv:2311.12793, 2023. 3
9. Liang Chen, Lei Li, Haozhe Zhao, Yifan Song, and Vinci. R1-v: Reinforcing super generalization ability in vision-language models with less than $3. [https://github.com/Deep-Agent/R1-V](https://github.com/Deep-Agent/R1-V), 2025. Accessed: 2025-02-02. 3, 4 [↗](https://github.com/Deep-Agent/R1-V)
10. Zhe Chen, Weiyun Wang, Yue Cao, Yangzhou Liu, Zhangwei Gao, Erfei Cui, Jinguo Zhu, Shenglong Ye, Hao Tian, Zhaoyang Liu, et al. [Expanding performance boundaries of open-source multimodal models with model, data, and test-time scaling](https://md.research.elamur.ai/free/en/papers/chen-2024-expanding-performance-boundaries-open/). arXiv preprint arXiv:2412.05271, 2024. 2, 3
11. Zhe Chen, Weiyun Wang, Hao Tian, Shenglong Ye, Zhangwei Gao, Erfei Cui, Wenwen Tong, Kongzhi Hu, Jiapeng Luo, Zheng Ma, et al. [How far are we to gpt-4v? closing the gap to commercial multimodal models with open-source suites](https://md.research.elamur.ai/free/en/papers/chen-2024-far-gpt-4v-closing/). arXiv preprint arXiv:2404.16821, 2024. 2, 3
12. Tianzhe Chu, Yuexiang Zhai, Jihan Yang, Shengbang Tong, Saining Xie, Dale Schuurmans, Quoc V Le, Sergey Levine, and Yi Ma. [Sft memorizes, rl generalizes: A comparative study of foundation model post-training](https://md.research.elamur.ai/free/en/papers/chu-2025-sft-memorizes-rl-generalizes-comparative/). arXiv preprint arXiv:2501.17161, 2025. 11
13. Wenliang Dai, Junnan Li, Dongxu Li, Anthony Meng Huat Tiong, Junqi Zhao, Weisheng Wang, Boyang Li, Pascale [find](https://scholar.google.com/scholar?q=Wenliang+Dai%2C+Junnan+Li%2C+Dongxu+Li%2C+Anthony+Meng+Huat+Tiong%2C+Junqi+Zhao%2C+Weisheng+Wang%2C+Boyang+Li%2C+Pascale)

*Updated 2 November 2026*