Preprint · 2025 · Open access

Diagnosing Visual Reasoning: Challenges, Insights, and a Path Forward

Bi, Jing, Guangyu Sun, Ali Vosoughi, Chen Chen, Chenliang Xu

arXiv (Cornell University) · Published 23 October 2025 · DOI 10.48550/arxiv.2510.20696

Open PDFView at the publisherCite this paper

In brief

“Diagnosing Visual Reasoning: Challenges, Insights, and a Path Forward” is a 2025 preprint by Bi, Jing and 4 co-authors, published in arXiv (Cornell University). OpenAlex counts 7 citations as of 19 November 2026. It is open access under a CC BY licence: the full text is free to read at arXiv (Cornell University).

Citations
7
Most cited in
2026 (5)
References
19
Authors
5

Abstract

Multimodal large language models (MLLMs) that integrate visual and textual reasoning leverage chain-of-thought (CoT) prompting to tackle complex visual tasks, yet continue to exhibit visual hallucinations and an over-reliance on textual priors. We present a systematic diagnosis of state-of-the-art vision-language models using a three-stage evaluation framework, uncovering key failure modes. To address these, we propose an agent-based architecture that combines LLM reasoning with lightweight visual modules, enabling fine-grained analysis and iterative refinement of reasoning chains. Our results highlight future visual reasoning models should focus on integrating a broader set of specialized tools for analyzing visual content. Our system achieves significant gains (+10.3 on MMMU, +6.0 on MathVista over a 7B baseline), matching or surpassing much larger models. We will release our framework and evaluation suite to facilitate future research.

Cite this paper

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

APA 7

Jing, B., Sun, G., Vosoughi, A., Chen, C., & Xu, C. (2025). Diagnosing Visual Reasoning: Challenges, Insights, and a Path Forward. arXiv (Cornell University). https://doi.org/10.48550/arxiv.2510.20696

MLA 9

Jing, Bi,, et al. “Diagnosing Visual Reasoning: Challenges, Insights, and a Path Forward.” arXiv (Cornell University), 2025. https://doi.org/10.48550/arxiv.2510.20696.

Chicago

Jing, Bi,, Guangyu Sun, Ali Vosoughi, Chen Chen, and Chenliang Xu. 2025. “Diagnosing Visual Reasoning: Challenges, Insights, and a Path Forward.” arXiv (Cornell University). https://doi.org/10.48550/arxiv.2510.20696.

Harvard

Jing, B. et al. (2025) ‘Diagnosing Visual Reasoning: Challenges, Insights, and a Path Forward’, arXiv (Cornell University). doi:10.48550/arxiv.2510.20696.

IEEE

B. Jing, G. Sun, A. Vosoughi, C. Chen, and C. Xu, “Diagnosing Visual Reasoning: Challenges, Insights, and a Path Forward,” arXiv (Cornell University), 2025, doi: 10.48550/arxiv.2510.20696.

BibTeX

@misc{jing2025diagnosing,
  title     = {{Diagnosing Visual Reasoning: Challenges, Insights, and a Path Forward}},
  author    = {Jing, Bi, and Sun, Guangyu and Vosoughi, Ali and Chen, Chen and Xu, Chenliang},
  howpublished = {arXiv (Cornell University)},
  year      = {2025},
  doi       = {10.48550/arxiv.2510.20696},
  url       = {https://doi.org/10.48550/arxiv.2510.20696}
}

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

  • visual reasoning
  • multimodal large language models
  • chain-of-thought prompting
  • visual hallucinations
  • failure mode identification
  • vision-language models
  • agent-based architecture
  • visual tools

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?

Jing, B., Sun, G., Vosoughi, A., Chen, C., & Xu, C. (2025). Diagnosing Visual Reasoning: Challenges, Insights, and a Path Forward. arXiv (Cornell University). https://doi.org/10.48550/arxiv.2510.20696

How many times has this paper been cited?

7 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?

Bi, Jing, Guangyu Sun, Ali Vosoughi, Chen Chen, Chenliang Xu.

What is the DOI of this paper?

10.48550/arxiv.2510.20696, permanently at https://doi.org/10.48550/arxiv.2510.20696. 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.