Journal article · 2026

Large language models are unstable as autonomous medical evaluators across clinical tasks

M Silva, Fernando Carvalho e Silva, Renata de Paula Gomes Nunes de Castro, Alex Goncalves, Artur Figueiredo Delgado

Journal of the American Medical Informatics Association · Published 28 October 2026 · DOI 10.1093/jamia/ocag190

View at the publisherCite this paper

In brief

“Large language models are unstable as autonomous medical evaluators across clinical tasks” is a 2026 journal article by M Silva and 4 co-authors, published in Journal of the American Medical Informatics Association. OpenAlex counts 0 citations as of 17 November 2026. The publisher's version is not open access, and OpenAlex records no free legal copy.

Citations
0
References
22
Authors
5

Abstract

OBJECTIVE: Large language models (LLMs) are increasingly used as automated evaluators of medical text ("LLM-as-judge"). We tested whether LLM-evaluator performance validated on one medical task against a human reference standard remains stable on a structurally different task. MATERIALS AND METHODS: We evaluated 10 LLMs (3 locally served open-source and 7 API models: Claude, GPT, DeepSeek)… Read the full abstract at the publisher.

Excerpt. This paper is not open access.

Cite this paper

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

APA 7

Silva, M., Silva, F. C. E., de Castro, R. D. P. G. N., Goncalves, A., & Delgado, A. F. (2026). Large language models are unstable as autonomous medical evaluators across clinical tasks. Journal of the American Medical Informatics Association. https://doi.org/10.1093/jamia/ocag190

MLA 9

Silva, M, et al. “Large language models are unstable as autonomous medical evaluators across clinical tasks.” Journal of the American Medical Informatics Association, 2026. https://doi.org/10.1093/jamia/ocag190.

Chicago

Silva, M, Fernando Carvalho e Silva, Renata de Paula Gomes Nunes de Castro, Alex Goncalves, and Artur Figueiredo Delgado. 2026. “Large language models are unstable as autonomous medical evaluators across clinical tasks.” Journal of the American Medical Informatics Association. https://doi.org/10.1093/jamia/ocag190.

Harvard

Silva, M. et al. (2026) ‘Large language models are unstable as autonomous medical evaluators across clinical tasks’, Journal of the American Medical Informatics Association. doi:10.1093/jamia/ocag190.

IEEE

M. Silva, F. C. E. Silva, R. D. P. G. N. de Castro, A. Goncalves, and A. F. Delgado, “Large language models are unstable as autonomous medical evaluators across clinical tasks,” Journal of the American Medical Informatics Association, 2026, doi: 10.1093/jamia/ocag190.

BibTeX

@article{silva2026large,
  title     = {{Large language models are unstable as autonomous medical evaluators across clinical tasks}},
  author    = {Silva, M and Silva, Fernando Carvalho e and de Castro, Renata de Paula Gomes Nunes and Goncalves, Alex and Delgado, Artur Figueiredo},
  journal   = {Journal of the American Medical Informatics Association},
  year      = {2026},
  doi       = {10.1093/jamia/ocag190},
  url       = {https://doi.org/10.1093/jamia/ocag190}
}

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

Topic and keywords

Health Sciences › Medicine › Health Informatics › Artificial Intelligence in Healthcare and Education

  • large language models
  • LLM-as-a-judge
  • performance stability
  • Emergency Severity Index
  • research appraisal
  • answer quality assessment
  • human oversight
  • clinical decision support

Work with this paper

Most-cited papers in Medicine

All highly cited Medicine papers

Questions about this paper

How do I cite this paper in APA style?

Silva, M., Silva, F. C. E., de Castro, R. D. P. G. N., Goncalves, A., & Delgado, A. F. (2026). Large language models are unstable as autonomous medical evaluators across clinical tasks. Journal of the American Medical Informatics Association. https://doi.org/10.1093/jamia/ocag190

How many times has this paper been cited?

0 times, according to OpenAlex on 17 November 2026. Google Scholar and Scopus count different sets of sources, so their numbers differ.

Is this paper free to read?

OpenAlex records no free legal copy. Check your library's access, the authors' pages or a repository, or ask the corresponding author for a copy.

Who wrote this paper?

M Silva, Fernando Carvalho e Silva, Renata de Paula Gomes Nunes de Castro, Alex Goncalves, Artur Figueiredo Delgado.

What is the DOI of this paper?

10.1093/jamia/ocag190, permanently at https://doi.org/10.1093/jamia/ocag190. Sign in and open it from this page.

Bibliographic data from OpenAlex (CC0), last checked 17 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.