Conference paper · 2025 · Open access

LlaVA-CoT: Let Vision Language Models Reason Step-By-Step

Guowei Xu, Peng Jin, Ziang Wu, Hao Li, Yibing Song, Lichao Sun, Li Yuan

IEEE/CVF International Conference on Computer Vision (ICCV), 2087–2098 · Published 19 October 2025 · DOI 10.1109/iccv51701.2025.00202

Open PDFView at the publisherCite this paper

In brief

“LlaVA-CoT: Let Vision Language Models Reason Step-By-Step” is a 2025 conference paper by Guowei Xu and 6 co-authors, published in IEEE/CVF International Conference on Computer Vision (ICCV). OpenAlex counts 655 citations as of 19 November 2026, which puts it in the top 1% of works published in 2025 by citations. It is open access: the full text is free to read at arXiv (Cornell University).

Citations
655
Most cited in
2026 (335)
Field-weighted impact
129.27
References
64
Authors
7

Abstract

Large language models have demonstrated substantial advancements in reasoning capabilities. However, current Vision-Language Models (VLMs) often struggle to perform systematic and structured reasoning, especially when handling complex visual question-answering tasks. In this work, we introduce LLaVA-CoT, a large VLM designed to conduct autonomous multistage reasoning. Unlike chain-of-thought prompting, LLaVA-CoT independently engages in sequential stages of summarization, visual interpretation, logical reasoning, and conclusion generation. This structured approach enables LLaVA-CoT to achieve marked improvements on reasoning-intensive tasks. To accomplish this, we construct the LLaVA-CoT-100k dataset, integrating samples from various visual question answering sources and providing structured reasoning annotations. Besides, we propose a test-time stage-wise retracing search method (SWIRES), which enables effective and efficient test-time scaling. Remarkably, with only 100k training samples and test-time scaling, LLaVA-CoT not only outperforms its base model by 9.4% on a wide range of multimodal reasoning benchmarks, but also surpasses the performance of larger and even closed-source models, such as Gemini-1.5-pro, GPT-4o-mini, and Llama-3.2-90B-Vision-Instruct. The code, dataset, and pre-trained weights are publicly available at https://github.com/PKU-YuanGroup/LLaVA-CoT.

Cite this paper

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

APA 7

Xu, G., Jin, P., Wu, Z., Li, H., Song, Y., Sun, L., & Yuan, L. (2025). LlaVA-CoT: Let Vision Language Models Reason Step-By-Step. IEEE/CVF International Conference on Computer Vision (ICCV), 2087–2098. https://doi.org/10.1109/iccv51701.2025.00202

MLA 9

Xu, Guowei, et al. “LlaVA-CoT: Let Vision Language Models Reason Step-By-Step.” IEEE/CVF International Conference on Computer Vision (ICCV), 2025, pp. 2087–2098. https://doi.org/10.1109/iccv51701.2025.00202.

Chicago

Xu, Guowei, Peng Jin, Ziang Wu, Hao Li, Yibing Song, Lichao Sun, and Li Yuan. 2025. “LlaVA-CoT: Let Vision Language Models Reason Step-By-Step.” IEEE/CVF International Conference on Computer Vision (ICCV): 2087–2098. https://doi.org/10.1109/iccv51701.2025.00202.

Harvard

Xu, G. et al. (2025) ‘LlaVA-CoT: Let Vision Language Models Reason Step-By-Step’, IEEE/CVF International Conference on Computer Vision (ICCV), pp. 2087–2098. doi:10.1109/iccv51701.2025.00202.

IEEE

G. Xu et al., “LlaVA-CoT: Let Vision Language Models Reason Step-By-Step,” IEEE/CVF International Conference on Computer Vision (ICCV), pp. 2087–2098, 2025, doi: 10.1109/iccv51701.2025.00202.

BibTeX

@inproceedings{xu2025llava,
  title     = {{LlaVA-CoT: Let Vision Language Models Reason Step-By-Step}},
  author    = {Xu, Guowei and Jin, Peng and Wu, Ziang and Li, Hao and Song, Yibing and Sun, Lichao and Yuan, Li},
  booktitle = {IEEE/CVF International Conference on Computer Vision (ICCV)},
  year      = {2025},
  pages     = {2087--2098},
  doi       = {10.1109/iccv51701.2025.00202},
  url       = {https://doi.org/10.1109/iccv51701.2025.00202}
}

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

Citations by year

YearCitationsShare of the best year
20241010
2025310310
2026335335

Its field-weighted citation impact is 129.27: it has been cited 129.27 times as often as the average work of the same field, type and year (1.00 is the world average).

Topic and keywords

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

  • vision-language models
  • chain-of-thought reasoning
  • visual question answering
  • multimodal reasoning
  • structured reasoning
  • test-time scaling
  • reasoning datasets

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?

Xu, G., Jin, P., Wu, Z., Li, H., Song, Y., Sun, L., & Yuan, L. (2025). LlaVA-CoT: Let Vision Language Models Reason Step-By-Step. IEEE/CVF International Conference on Computer Vision (ICCV), 2087–2098. https://doi.org/10.1109/iccv51701.2025.00202

How many times has this paper been cited?

655 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?

Guowei Xu, Peng Jin, Ziang Wu, Hao Li, Yibing Song, Lichao Sun, Li Yuan.

What is the DOI of this paper?

10.1109/iccv51701.2025.00202, permanently at https://doi.org/10.1109/iccv51701.2025.00202. 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.