Preprint · 2025 · Open access

VGR: Visual Grounded Reasoning

Wang, Jiacong, Kang, Zijian, Haochen Wang, Jiang, Haiyong, Jiawen Li, Bohong Wu, Yajun Wang, Ran, Jiao, Xiao Qing Liang, Chao Qing Feng, Jun Xiao

arXiv (Cornell University) · Published 13 June 2025 · DOI 10.48550/arxiv.2506.11991

Open PDFView at the publisherCite this paper

In brief

“VGR: Visual Grounded Reasoning” is a 2025 preprint by Wang, Jiacong and 10 co-authors, published in arXiv (Cornell University). OpenAlex counts 89 citations as of 19 November 2026. It is open access: the full text is free to read at arXiv (Cornell University).

Citations
89
Most cited in
2026 (57)
References
53
Authors
11

Abstract

In the field of multimodal chain-of-thought (CoT) reasoning, existing approaches predominantly rely on reasoning on pure language space, which inherently suffers from language bias and is largely confined to math or science domains. This narrow focus limits their ability to handle complex visual reasoning tasks that demand comprehensive understanding of image details. To address these limitations, this paper introduces VGR, a novel reasoning multimodal large language model (MLLM) with enhanced fine-grained visual perception capabilities. Unlike traditional MLLMs that answer the question or reasoning solely on the language space, our VGR first detects relevant regions that may help to solve problems, and then provides precise answers based on replayed image regions. To achieve this, we conduct a large-scale SFT dataset called VGR -SFT that contains reasoning data with mixed vision grounding and language deduction. The inference pipeline of VGR allows the model to choose bounding boxes for visual reference and a replay stage is introduced to integrates the corresponding regions into the reasoning process, enhancing multimodel comprehension. Experiments on the LLaVA-NeXT-7B baseline show that VGR achieves superior performance on multi-modal benchmarks requiring comprehensive image detail understanding. Compared to the baseline, VGR uses only 30\% of the image token count while delivering scores of +4.1 on MMStar, +7.1 on AI2D, and a +12.9 improvement on ChartQA.

Cite this paper

Formatted from the catalogue record. Check names, capitals and page numbers against the paper before you submit.

APA 7

Jiacong, W., Zijian, K., Wang, H., Haiyong, J., Li, J., Wu, B., Wang, Y., Jiao, R., Liang, X. Q., Feng, C. Q., & Xiao, J. (2025). VGR: Visual Grounded Reasoning. arXiv (Cornell University). https://doi.org/10.48550/arxiv.2506.11991

MLA 9

Jiacong, Wang,, et al. “VGR: Visual Grounded Reasoning.” arXiv (Cornell University), 2025. https://doi.org/10.48550/arxiv.2506.11991.

Chicago

Jiacong, Wang,, Kang, Zijian, Haochen Wang, Jiang, Haiyong, Jiawen Li, Bohong Wu, Yajun Wang, et al. 2025. “VGR: Visual Grounded Reasoning.” arXiv (Cornell University). https://doi.org/10.48550/arxiv.2506.11991.

Harvard

Jiacong, W. et al. (2025) ‘VGR: Visual Grounded Reasoning’, arXiv (Cornell University). doi:10.48550/arxiv.2506.11991.

IEEE

W. Jiacong et al., “VGR: Visual Grounded Reasoning,” arXiv (Cornell University), 2025, doi: 10.48550/arxiv.2506.11991.

BibTeX

@misc{jiacong2025visual,
  title     = {{VGR: Visual Grounded Reasoning}},
  author    = {Jiacong, Wang, and Zijian, Kang, and Wang, Haochen and Haiyong, Jiang, and Li, Jiawen and Wu, Bohong and Wang, Yajun and Jiao, Ran, and Liang, Xiao Qing and Feng, Chao Qing and Xiao, Jun},
  howpublished = {arXiv (Cornell University)},
  year      = {2025},
  doi       = {10.48550/arxiv.2506.11991},
  url       = {https://doi.org/10.48550/arxiv.2506.11991}
}

Download for a reference manager: BibTeX (.bib) · RIS for Zotero, Mendeley and EndNote (.ris)

Topic and keywords

Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Multimodal Machine Learning Applications

  • multimodal large language models
  • chain-of-thought reasoning
  • language bias

Work with this paper

Most-cited papers in Computer Science

All highly cited Computer Science papers

Questions about this paper

How do I cite this paper in APA style?

Jiacong, W., Zijian, K., Wang, H., Haiyong, J., Li, J., Wu, B., Wang, Y., Jiao, R., Liang, X. Q., Feng, C. Q., & Xiao, J. (2025). VGR: Visual Grounded Reasoning. arXiv (Cornell University). https://doi.org/10.48550/arxiv.2506.11991

How many times has this paper been cited?

89 times, according to OpenAlex on 19 November 2026. Google Scholar and Scopus count different sets of sources, so their numbers differ.

Is this paper free to read?

Yes. It is open access at arXiv (Cornell University): sign in and open the PDF from this page.

Who wrote this paper?

Wang, Jiacong, Kang, Zijian, Haochen Wang, Jiang, Haiyong, Jiawen Li, Bohong Wu, Yajun Wang, Ran, Jiao, Xiao Qing Liang, Chao Qing Feng, Jun Xiao.

What is the DOI of this paper?

10.48550/arxiv.2506.11991, permanently at https://doi.org/10.48550/arxiv.2506.11991. Sign in and open it from this page.

Bibliographic data from OpenAlex (CC0), last checked 19 November 2026. ResearchGuru does not host this paper; once you are signed in, the links open it at the publisher or the repository that does.