Journal article · 2026 · Open access

Scaling medical imaging report generation with multimodal reinforcement learning

Qianchu Liu, Sheng Zhang, Guanghui Qin, Yu Gu, Ying Jin, Sam Preston, Yanbo Xu, Sid Kiblawi, Wen-wai Yim, Timothy Ossowski, Tristan Naumann, Mu Wei and 1 more

Nature Communications · Published 12 November 2026 · DOI 10.1038/s41467-026-78767-4

Open PDFView at the publisherCite this paper

In brief

“Scaling medical imaging report generation with multimodal reinforcement learning” is a 2026 journal article by Qianchu Liu and 12 co-authors, published in Nature Communications. OpenAlex counts 14 citations as of 18 November 2026, which puts it in the top 2% of works published in 2026 by citations. It is open access under a CC BY licence: the full text is free to read at Nature Communications.

Citations
14
Most cited in
2026 (14)
Field-weighted impact
17.62
References
31
Authors
13

Abstract

In medical imaging report generation, supervised fine-tuning can overfit to superficial patterns and generalize poorly across institutions. We introduce Universal Report Generation (UNIRG), which adopts reinforcement learning to directly optimize clinically meaningful quality criteria. With this framework, we developed UNIRG-CXR, an 8-billion-parameter vision–language model trained on four public datasets (MIMIC-CXR, CheXpert Plus, ReXGradient-160k and IU-Xray) by supervised fine-tuning followed by reinforcement learning (group-relative policy optimization) against a composite lexical, semantic and clinical-error reward. We evaluated held-out test sets and an external proprietary dataset, with RadCliQ as primary metric and expert radiologist validation. Here we show that our method improves report quality and generalization: UNIRG-CXR surpasses prior models on the ReXrank benchmark across four datasets and two generation settings, exceeding the previous best score by over 50% on two test sets, and produces fewer clinically significant errors while maintaining performance on unseen data sources and across demographic subgroups.

Cite this paper

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

APA 7

Liu, Q., Zhang, S., Qin, G., Gu, Y., Jin, Y., Preston, S., Xu, Y., Kiblawi, S., Yim, W.-W., Ossowski, T., Naumann, T., Wei, M., & Poon, H. (2026). Scaling medical imaging report generation with multimodal reinforcement learning. Nature Communications. https://doi.org/10.1038/s41467-026-78767-4

MLA 9

Liu, Qianchu, et al. “Scaling medical imaging report generation with multimodal reinforcement learning.” Nature Communications, 2026. https://doi.org/10.1038/s41467-026-78767-4.

Chicago

Liu, Qianchu, Sheng Zhang, Guanghui Qin, Yu Gu, Ying Jin, Sam Preston, Yanbo Xu, et al. 2026. “Scaling medical imaging report generation with multimodal reinforcement learning.” Nature Communications. https://doi.org/10.1038/s41467-026-78767-4.

Harvard

Liu, Q. et al. (2026) ‘Scaling medical imaging report generation with multimodal reinforcement learning’, Nature Communications. doi:10.1038/s41467-026-78767-4.

IEEE

Q. Liu et al., “Scaling medical imaging report generation with multimodal reinforcement learning,” Nature Communications, 2026, doi: 10.1038/s41467-026-78767-4.

BibTeX

@article{liu2026scaling,
  title     = {{Scaling medical imaging report generation with multimodal reinforcement learning}},
  author    = {Liu, Qianchu and Zhang, Sheng and Qin, Guanghui and Gu, Yu and Jin, Ying and Preston, Sam and Xu, Yanbo and Kiblawi, Sid and Yim, Wen-wai and Ossowski, Timothy and Naumann, Tristan and Wei, Mu and Poon, Hoifung},
  journal   = {Nature Communications},
  year      = {2026},
  doi       = {10.1038/s41467-026-78767-4},
  url       = {https://doi.org/10.1038/s41467-026-78767-4}
}

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

  • chest X-ray reports
  • vision-language models
  • policy optimization
  • radiology report quality

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?

Liu, Q., Zhang, S., Qin, G., Gu, Y., Jin, Y., Preston, S., Xu, Y., Kiblawi, S., Yim, W.-W., Ossowski, T., Naumann, T., Wei, M., & Poon, H. (2026). Scaling medical imaging report generation with multimodal reinforcement learning. Nature Communications. https://doi.org/10.1038/s41467-026-78767-4

How many times has this paper been cited?

14 times, according to OpenAlex on 18 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 Nature Communications: sign in and open the PDF from this page.

Who wrote this paper?

Qianchu Liu, Sheng Zhang, Guanghui Qin, Yu Gu, Ying Jin, Sam Preston, Yanbo Xu, Sid Kiblawi, Wen-wai Yim, Timothy Ossowski, Tristan Naumann, Mu Wei and 1 more.

What is the DOI of this paper?

10.1038/s41467-026-78767-4, permanently at https://doi.org/10.1038/s41467-026-78767-4. Sign in and open it from this page.

Bibliographic data from OpenAlex (CC0), last checked 18 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.