{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,22]],"date-time":"2026-10-22T06:37:52Z","timestamp":1792651072496,"version":"4.5.0"},"reference-count":73,"publisher":"Springer Science and Business Media LLC","issue":"1","license":[{"start":{"date-parts":[[2026,8,29]],"date-time":"2026-08-29T00:00:00Z","timestamp":1787961600000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2026,10,22]],"date-time":"2026-10-22T00:00:00Z","timestamp":1792627200000},"content-version":"vor","delay-in-days":54,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["Commun Med"],"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:sec>\n                    <jats:title>Background<\/jats:title>\n                    <jats:p>Large language models (LLMs) are emerging as tools to support clinical decision making. HIV management is a compelling use case due to its complexity and dynamic nature, involving diverse treatment options, comorbidities, and adherence challenges. However, integrating LLMs into clinical practice raises concerns about accuracy, safety, and clinician acceptance. Despite growing interest, their performance in HIV care remains poorly studied, and benchmarking is lacking.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>We developed HIVMedQA, a clinician-curated benchmark of HIV-related open-ended medical question-answer pairs spanning basic knowledge, clinical reasoning, complex patient vignettes, and bias-modified scenarios. We evaluated seven general-purpose and three medical LLMs. Performance was assessed using lexical similarity and an extended medical LLM-as-a-judge framework capturing key clinical dimensions, including question comprehension, reasoning, knowledge recall, bias, potential harm, and factual accuracy, to better capture nuances relevant to the medical domain, with additional evaluation by HIV-experienced physicians.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>Performance varies substantially across models and task complexity. Gemini 2.5 Pro achieves the highest overall scores, followed by Claude 3.5 Sonnet and MedGemma-27B. Knowledge recall is generally stronger than question comprehension or clinical reasoning. Medical LLMs do not consistently outperform general-purpose models, and model size alone does not predict performance. Several models are sensitive to cognitive bias prompts. LLM-as-a-judge scoring aligns better with clinician assessment than lexical metrics.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusions<\/jats:title>\n                    <jats:p>HIVMedQA provides a structured benchmark for evaluating LLMs in HIV clinical decision support. Current LLMs show promise, but limitations in reasoning, bias robustness, and safety indicate that careful validation, domain-specific evaluation, and clinician oversight remain essential before clinical deployment.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1038\/s43856-026-01875-1","type":"journal-article","created":{"date-parts":[[2026,8,29]],"date-time":"2026-08-29T18:33:57Z","timestamp":1788028437000},"update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":1,"title":["Benchmarking large language models for HIV medical decision support"],"prefix":"10.1038","volume":"6","author":[{"given":"Gonzalo","family":"Cardenal-Antolin","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-8240-939X","authenticated-orcid":false,"given":"Jacques","family":"Fellay","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0005-4095-0152","authenticated-orcid":false,"given":"Bashkim","family":"Jaha","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-1142-6723","authenticated-orcid":false,"given":"Huldrych F.","family":"G\u00fcnthard","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Johannes","family":"Nemeth","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-9220-8348","authenticated-orcid":false,"given":"Roger","family":"Kouyos","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-0573-6119","authenticated-orcid":false,"given":"Niko","family":"Beerenwinkel","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-8564-3157","authenticated-orcid":false,"given":"Diane","family":"Duroux","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"297","published-online":{"date-parts":[[2026,8,29]]},"reference":[{"key":"1875_CR1","doi-asserted-by":"publisher","first-page":"457","DOI":"10.1016\/S0140-6736(25)00202-8","volume":"405","author":"J Zou","year":"2025","unstructured":"Zou, J. & Topol, E. J. The rise of agentic AI teammates in medicine. The Lancet 405, 457 (2025).","journal-title":"The Lancet"},{"key":"1875_CR2","doi-asserted-by":"crossref","unstructured":"Clough, R. A. J. et al. Transforming healthcare documentation: harnessing the potential of AI to generate discharge summaries. BJGP Open 8, BJGPO.2023.0116 (2024).","DOI":"10.3399\/BJGPO.2023.0116"},{"key":"1875_CR3","doi-asserted-by":"publisher","first-page":"e101120","DOI":"10.1136\/bmjhci-2024-101120","volume":"31","author":"S Pavuluri","year":"2024","unstructured":"Pavuluri, S., Sangal, R., Sather, J. & Taylor, R. A. Balancing act: the complex role of artificial intelligence in addressing burnout and healthcare workforce dynamics. BMJ Health & Care Informatics 31, e101120 (2024).","journal-title":"BMJ Health & Care Informatics"},{"key":"1875_CR4","doi-asserted-by":"publisher","first-page":"e107","DOI":"10.1016\/S2589-7500(23)00021-3","volume":"5","author":"SB Patel","year":"2023","unstructured":"Patel, S. B. & Lam, K. Chatgpt: the future of discharge summaries? The Lancet Digital Health 5, e107\u2013e108 (2023).","journal-title":"The Lancet Digital Health"},{"key":"1875_CR5","doi-asserted-by":"crossref","unstructured":"Meng, X. et al. The application of large language models in medicine: a scoping review. Iscience 27, 109713 (2024).","DOI":"10.1016\/j.isci.2024.109713"},{"key":"1875_CR6","doi-asserted-by":"publisher","first-page":"1930","DOI":"10.1038\/s41591-023-02448-8","volume":"29","author":"AJ Thirunavukarasu","year":"2023","unstructured":"Thirunavukarasu, A. J. et al. Large language models in medicine. Nature Medicine 29, 1930\u20131940 (2023).","journal-title":"Nature Medicine"},{"key":"1875_CR7","doi-asserted-by":"publisher","DOI":"10.1186\/s12909-024-06399-7","volume":"24","author":"E Br\u00fcgge","year":"2024","unstructured":"Br\u00fcgge, E. et al. Large language models improve clinical decision making of medical students through patient simulation and structured feedback: a randomized controlled trial. BMC Medical Education 24, 1391 (2024).","journal-title":"BMC Medical Education"},{"key":"1875_CR8","doi-asserted-by":"crossref","unstructured":"Wang, S. et al. Capabilities of gpt-5 on multimodal medical reasoning. In Medical Imaging 2026: Imaging Informatics, (eds. Yang, X. & Hsu, W.) Vol. 13930, 139300E. International Society for Optics and Photonics (SPIE, 2026).","DOI":"10.1117\/12.3086794"},{"key":"1875_CR9","doi-asserted-by":"publisher","first-page":"1233","DOI":"10.1056\/NEJMsr2214184","volume":"388","author":"P Lee","year":"2023","unstructured":"Lee, P., Bubeck, S. & Petro, J. Benefits, limits, and risks of GPT-4 as an AI chatbot for medicine. New England Journal of Medicine 388, 1233\u20131239 (2023).","journal-title":"New England Journal of Medicine"},{"key":"1875_CR10","doi-asserted-by":"crossref","unstructured":"Nachane, S. S. et al. Few shot chain-of-thought driven reasoning to prompt LLMs for open-ended medical question answering. In Findings of the Association for Computational Linguistics: EMNLP 2024, (eds. Al-Onaizan, Y., Bansal, M. & Chen, Y.-N.) 542\u2013573 (Association for Computational Linguistics, 2024).","DOI":"10.18653\/v1\/2024.findings-emnlp.31"},{"key":"1875_CR11","doi-asserted-by":"crossref","unstructured":"Li, Y. et al. Chatdoctor: a medical chat model fine-tuned on a large language model meta-ai (Llama) using medical domain knowledge. Cureus 15, e40895 (2023).","DOI":"10.7759\/cureus.40895"},{"key":"1875_CR12","doi-asserted-by":"publisher","first-page":"e017902","DOI":"10.1136\/bmjopen-2017-017902","volume":"7","author":"G Irving","year":"2017","unstructured":"Irving, G. et al. International variations in primary care physician consultation time: a systematic review of 67 countries. BMJ Open 7, e017902 (2017).","journal-title":"BMJ Open"},{"key":"1875_CR13","doi-asserted-by":"publisher","first-page":"004695802091030","DOI":"10.1177\/0046958020910305","volume":"57","author":"D McIntyre","year":"2020","unstructured":"McIntyre, D. & Chow, C. K. Waiting time as an indicator for health services under strain: a narrative review. Inquiry 57, 0046958020910305 (2020).","journal-title":"Inquiry"},{"key":"1875_CR14","doi-asserted-by":"publisher","first-page":"109","DOI":"10.2471\/BLT.19.234138","volume":"98","author":"G Cometto","year":"2019","unstructured":"Cometto, G., Buchan, J. & Dussault, G. Developing the health workforce for universal health coverage. Bulletin of the World Health Organization 98, 109 (2019).","journal-title":"Bulletin of the World Health Organization"},{"key":"1875_CR15","doi-asserted-by":"publisher","first-page":"e2457879","DOI":"10.1001\/jamanetworkopen.2024.57879","volume":"8","author":"B Huo","year":"2025","unstructured":"Huo, B. et al. Large language models for chatbot health advice studies: a systematic review. JAMA Network Open 8, e2457879\u2013e2457879 (2025).","journal-title":"JAMA Network Open"},{"key":"1875_CR16","doi-asserted-by":"publisher","first-page":"860","DOI":"10.1093\/cid\/ciad633","volume":"78","author":"IS Schwartz","year":"2024","unstructured":"Schwartz, I. S., Link, K. E., Daneshjou, R. & Cort\u00e9s-Penfield, N. Black box warning: large language models and the future of infectious diseases consultation. Clinical Infectious Diseases 78, 860\u2013866 (2024).","journal-title":"Clinical Infectious Diseases"},{"key":"1875_CR17","doi-asserted-by":"publisher","DOI":"10.1038\/s41746-023-00939-z","volume":"6","author":"JA Omiye","year":"2023","unstructured":"Omiye, J. A., Lester, J. C., Spichak, S., Rotemberg, V. & Daneshjou, R. Large language models propagate race-based medicine. NPJ Digital Medicine 6, 195 (2023).","journal-title":"NPJ Digital Medicine"},{"key":"1875_CR18","doi-asserted-by":"publisher","first-page":"176","DOI":"10.1038\/s43856-024-00601-z","volume":"4","author":"Y Yang","year":"2024","unstructured":"Yang, Y., Liu, X., Jin, Q., Huang, F. & Lu, Z. Unmasking and quantifying racial bias of large language models in medical report generation. Communications Medicine 4, 176 (2024).","journal-title":"Communications Medicine"},{"key":"1875_CR19","unstructured":"Kanithi, P. K. et al. Medic: towards a comprehensive framework for evaluating llms in clinical applications. Preprint at https:\/\/arxiv.org\/abs\/2409.07314v1 (2024)."},{"key":"1875_CR20","doi-asserted-by":"publisher","first-page":"589","DOI":"10.1001\/jamainternmed.2023.1838","volume":"183","author":"JW Ayers","year":"2023","unstructured":"Ayers, J. W. et al. Comparing physician and artificial intelligence chatbot responses to patient questions posted to a public social media forum. JAMA Internal Medicine 183, 589\u2013596 (2023).","journal-title":"JAMA Internal Medicine"},{"key":"1875_CR21","doi-asserted-by":"crossref","unstructured":"Schmidgall, S. et al. Agentclinic: a multimodal benchmark for tool-using clinical ai agents. npj Digital Medicine 9, 499 (2026).","DOI":"10.1038\/s41746-026-02674-7"},{"key":"1875_CR22","doi-asserted-by":"crossref","unstructured":"Johri, S. et al. An evaluation framework for clinical use of large language models in patient interaction tasks. Nat. Med. 31, 1\u201310 (2025).","DOI":"10.1038\/s41591-024-03328-5"},{"key":"1875_CR23","unstructured":"Chen, Z. et al. Meditron-70b: scaling medical pretraining for large language models. Preprint at https:\/\/arxiv.org\/abs\/2311.16079 (2023)."},{"key":"1875_CR24","doi-asserted-by":"publisher","first-page":"1090","DOI":"10.1227\/neu.0000000000002551","volume":"93","author":"R Ali","year":"2023","unstructured":"Ali, R. et al. Performance of ChatGPT, GPT-4, and Google Bard on a neurosurgery oral boards preparation question bank. Neurosurgery 93, 1090\u20131098 (2023).","journal-title":"Neurosurgery"},{"key":"1875_CR25","doi-asserted-by":"publisher","first-page":"e0000198","DOI":"10.1371\/journal.pdig.0000198","volume":"2","author":"TH Kung","year":"2023","unstructured":"Kung, T. H. et al. Performance of ChatGPT on USMLE: potential for AI-assisted medical education using large language models. PLoS Digital Health 2, e0000198 (2023).","journal-title":"PLoS Digital Health"},{"key":"1875_CR26","doi-asserted-by":"crossref","unstructured":"Wang, C. et al. Evaluating open-qa evaluation. Adv. Neural Inf. Process. Syst. 36, 77013\u201377042 (2023).","DOI":"10.52202\/075280-3367"},{"key":"1875_CR27","unstructured":"Nori, H. et al. Can generalist foundation models outcompete special-purpose tuning? case study in medicine. Preprint at https:\/\/arxiv.org\/abs\/2311.16452 (2023)."},{"key":"1875_CR28","doi-asserted-by":"publisher","first-page":"1015","DOI":"10.1093\/jamia\/ocaf045","volume":"32","author":"FJ Dorfner","year":"2025","unstructured":"Dorfner, F. J. et al. Evaluating the effectiveness of biomedical fine-tuning for large language models on clinical tasks. Journal of the American Medical Informatics Association 32, 1015\u20131024 (2025).","journal-title":"Journal of the American Medical Informatics Association"},{"key":"1875_CR29","doi-asserted-by":"publisher","first-page":"e46599","DOI":"10.2196\/46599","volume":"9","author":"AJ Thirunavukarasu","year":"2023","unstructured":"Thirunavukarasu, A. J. et al. Trialling a large language model (ChatGPT) in general practice with the applied knowledge test: observational study demonstrating opportunities and limitations in primary care. JMIR Medical Education 9, e46599 (2023).","journal-title":"JMIR Medical Education"},{"key":"1875_CR30","doi-asserted-by":"publisher","first-page":"56","DOI":"10.9734\/ISRR\/2023\/v12i2164","volume":"12","author":"H Singh","year":"2023","unstructured":"Singh, H. et al. Navigating complexities in HIV care: challenges, solutions, and strategies. International STD Research & Reviews 12, 56\u201362 (2023).","journal-title":"International STD Research & Reviews"},{"key":"1875_CR31","doi-asserted-by":"publisher","first-page":"42","DOI":"10.1038\/s41572-023-00452-3","volume":"9","author":"L-G Bekker","year":"2023","unstructured":"Bekker, L.-G. et al. Hiv infection. Nature Reviews Disease Primers 9, 42 (2023).","journal-title":"Nature Reviews Disease Primers"},{"key":"1875_CR32","doi-asserted-by":"publisher","first-page":"4961","DOI":"10.1007\/s12325-021-01883-8","volume":"38","author":"GA McComsey","year":"2021","unstructured":"McComsey, G. A., Lingohr-Smith, M., Rogers, R., Lin, J. & Donga, P. Real-world adherence to antiretroviral therapy among HIV-1 patients across the United States. Advances in therapy 38, 4961\u20134974 (2021).","journal-title":"Advances in therapy"},{"key":"1875_CR33","doi-asserted-by":"publisher","first-page":"404","DOI":"10.1093\/cid\/ciae406","volume":"80","author":"D Rupasinghe","year":"2025","unstructured":"Rupasinghe, D. et al. Integrase strand transfer inhibitor\u2013related changes in body mass index and risk of diabetes: a prospective study from the Respondent Cohort Consortium. Clinical Infectious Diseases 80, 404\u2013416 (2025).","journal-title":"Clinical Infectious Diseases"},{"key":"1875_CR34","unstructured":"Dubey, A. et al. The llama 3 herd of models. Preprint at https:\/\/arxiv.org\/abs\/2407.21783 (2024)."},{"key":"1875_CR35","unstructured":"Dai, W. et al. Nvlm: open frontier-class multimodal LLMs. Preprint at https:\/\/arxiv.org\/abs\/2409.11402 (2024)."},{"key":"1875_CR36","unstructured":"Christophe, C., Kanithi, P. K., Raha, T., Khan, S. & Pimentel, M. A. Med42-v2: a suite of clinical LLMs. Preprint at https:\/\/arxiv.org\/abs\/2408.06142 (2024)."},{"key":"1875_CR37","doi-asserted-by":"publisher","first-page":"6421","DOI":"10.3390\/app11146421","volume":"11","author":"D Jin","year":"2021","unstructured":"Jin, D. et al. What disease does this patient have? A large-scale open domain question answering dataset from medical exams. Applied Sciences 11, 6421 (2021).","journal-title":"Applied Sciences"},{"key":"1875_CR38","unstructured":"Jin, D. et al. MedQA Dataset; https:\/\/huggingface.co\/datasets\/bigbio\/med_qa hosted on Hugging Face.(2021). Accessed 9 December 9 2024."},{"key":"1875_CR39","doi-asserted-by":"publisher","DOI":"10.1038\/s41746-024-01283-6","volume":"7","author":"S Schmidgall","year":"2024","unstructured":"Schmidgall, S. et al. Evaluation and mitigation of cognitive biases in medical language models. npj Digital Medicine 7, 295 (2024).","journal-title":"npj Digital Medicine"},{"key":"1875_CR40","doi-asserted-by":"crossref","unstructured":"Yao, Z. et al. Medqa-cs: objective structured clinical examination (OSCE)-style benchmark for evaluating LLM clinical skills. In Proceedings of the 19th Conference of the European Chapter of the Association for Computational Linguistics (Volume 1: Long Papers), (eds. Demberg, V., Inui, K. & Marquez, L.) 6183\u20136257 (Association for Computational Linguistics, 2026).","DOI":"10.18653\/v1\/2026.eacl-long.292"},{"key":"1875_CR41","doi-asserted-by":"publisher","first-page":"124575","DOI":"10.52202\/079017-3958","volume":"37","author":"S Kweon","year":"2024","unstructured":"Kweon, S. et al. Ehrnoteqa: an LLM benchmark for real-world clinical practice using discharge summaries. Advances in Neural Information Processing Systems 37, 124575\u2013124611 (2024).","journal-title":"Advances in Neural Information Processing Systems"},{"key":"1875_CR42","unstructured":"Hosseini, P. et al. A benchmark for long-form medical question answering. Preprint at https:\/\/arxiv.org\/abs\/2411.09834 (2024)."},{"key":"1875_CR43","doi-asserted-by":"crossref","unstructured":"Wang, Z. et al. Healthq: unveiling questioning capabilities of LLM chains in healthcare conversations. Smart Health 36, 100570 (2025).","DOI":"10.1016\/j.smhl.2025.100570"},{"key":"1875_CR44","unstructured":"Wang, J., Yang, Z., Yao, Z. & Yu, H. Jmlr: Joint medical llm and retrieval training for enhancing reasoning and professional question answering capability. arXiv Preprint at https:\/\/arxiv.org\/abs\/2402.17887 (2024)."},{"key":"1875_CR45","doi-asserted-by":"crossref","unstructured":"Fu, J., Ng, S. K., Jiang, Z. & Liu, P. Gptscore: evaluate as you desire. In Proceedings of the 2024 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers), (eds. Duh, K., Gomez, H. & Bethard, S.) 6556\u20136576 (Association for Computational Linguistics, 2024).","DOI":"10.18653\/v1\/2024.naacl-long.365"},{"key":"1875_CR46","doi-asserted-by":"publisher","first-page":"172","DOI":"10.1038\/s41586-023-06291-2","volume":"620","author":"K Singhal","year":"2023","unstructured":"Singhal, K. et al. Large language models encode clinical knowledge. Nature 620, 172\u2013180 (2023).","journal-title":"Nature"},{"key":"1875_CR47","unstructured":"Lin, C.-Y. Rouge: a package for automatic evaluation of summaries. In Text Summarisation Branches Out, 74\u201381 (Association for Computational Linguistics, 2004)."},{"key":"1875_CR48","doi-asserted-by":"crossref","unstructured":"Papineni, K., Roukos, S., Ward, T. & Zhu, W.-J. Bleu: a method for automatic evaluation of machine translation. In Proceedings of the 40th annual meeting of the Association for Computational Linguistics, (eds. Isabelle, P., Charniak, E. & Lin, D.) 311\u2013318 (Association for Computational Linguistics, 2002).","DOI":"10.3115\/1073083.1073135"},{"key":"1875_CR49","doi-asserted-by":"publisher","first-page":"D267","DOI":"10.1093\/nar\/gkh061","volume":"32","author":"O Bodenreider","year":"2004","unstructured":"Bodenreider, O. The unified medical language system (UMLS): integrating biomedical terminology. Nucleic Acids Research 32, D267\u2013D270 (2004).","journal-title":"Nucleic Acids Research"},{"key":"1875_CR50","doi-asserted-by":"crossref","unstructured":"Neumann, M., King, D., Beltagy, I. & Ammar, W. ScispaCy: Fast and robust models for biomedical natural language processing. In Proceedings of the 18th BioNLP Workshop and Shared Task, (eds. Demner-Fushman, D., Cohen, K. B., Ananiadou, S. & Tsujii, J.) 319\u2013327 (Association for Computational Linguistics, 2019).","DOI":"10.18653\/v1\/W19-5034"},{"key":"1875_CR51","unstructured":"Lamy, J.-B., Venot, A. & Duclos, C. Pymedtermino: an open-source generic api for advanced terminology services. In Digital Healthcare Empowering Europeans, (eds. Cornet, R. et al.) 924\u2013928 (IOS Press, 2015)."},{"key":"1875_CR52","doi-asserted-by":"publisher","first-page":"11","DOI":"10.1016\/j.artmed.2017.07.002","volume":"80","author":"J-B Lamy","year":"2017","unstructured":"Lamy, J.-B. Owlready: ontology-oriented programming in Python with automatic classification and high level constructs for biomedical ontologies. Artificial Intelligence in Medicine 80, 11\u201328 (2017).","journal-title":"Artificial Intelligence in Medicine"},{"key":"1875_CR53","unstructured":"National Library of Medicine (US). Unified Medical Language System (UMLS): 2024AB Full Release Files; https:\/\/www.nlm.nih.gov\/research\/umls\/licensedcontent\/umlsknowledgesources.html (2024). Accessed 16 December 2024."},{"key":"1875_CR54","doi-asserted-by":"crossref","unstructured":"Bird, S. Nltk: the natural language toolkit. In Proceedings of the COLING\/ACL 2006 Interactive Presentation Sessions, (ed. Curran, J.) 69\u201372 (Association for Computational Linguistics, 2006).","DOI":"10.3115\/1225403.1225421"},{"key":"1875_CR55","doi-asserted-by":"publisher","first-page":"39","DOI":"10.1145\/219717.219748","volume":"38","author":"GA Miller","year":"1995","unstructured":"Miller, G. A. Wordnet: a lexical database for English. Communications of the ACM 38, 39\u201341 (1995).","journal-title":"Communications of the ACM"},{"key":"1875_CR56","doi-asserted-by":"publisher","first-page":"1892","DOI":"10.1093\/jamia\/ocab090","volume":"28","author":"Y Zhang","year":"2021","unstructured":"Zhang, Y., Zhang, Y., Qi, P., Manning, C. D. & Langlotz, C. P. Biomedical and clinical English model packages for the Stanza Python NLP Library. Journal of the American Medical Informatics Association 28, 1892\u20131899 (2021).","journal-title":"Journal of the American Medical Informatics Association"},{"key":"1875_CR57","doi-asserted-by":"publisher","first-page":"S43","DOI":"10.1046\/j.1523-1755.63.s83.10.x","volume":"63","author":"JB Kopp","year":"2003","unstructured":"Kopp, J. B. & Winkler, C. HIV-associated nephropathy in African Americans. Kidney International 63, S43\u2013S49 (2003).","journal-title":"Kidney International"},{"key":"1875_CR58","doi-asserted-by":"publisher","first-page":"2129","DOI":"10.1681\/ASN.2011040388","volume":"22","author":"JB Kopp","year":"2011","unstructured":"Kopp, J. B. et al. Apol1 genetic variants in focal segmental glomerulosclerosis and HIV-associated nephropathy. Journal of the American Society of Nephrology 22, 2129\u20132137 (2011).","journal-title":"Journal of the American Society of Nephrology"},{"key":"1875_CR59","doi-asserted-by":"publisher","unstructured":"Choi, W. C. & Chang, C. I. Advantages and limitations of open-source versus commercial large language models (LLMs): a comparative study of DeepSeek and OpenAI\u2019s ChatGPT. Preprint at https:\/\/doi.org\/10.20944\/preprints202503.1081.v1 (2025).","DOI":"10.20944\/preprints202503.1081.v1"},{"key":"1875_CR60","doi-asserted-by":"crossref","unstructured":"Manchanda, J., Boettcher, L., Westphalen, M. & Jasser, J. The open-source advantage in large language models (llms). AI and Ethics 6, 390 (2026).","DOI":"10.1007\/s43681-026-01190-4"},{"key":"1875_CR61","unstructured":"Team, G. et al. Gemini: a family of highly capable multimodal models. Preprint at https:\/\/arxiv.org\/abs\/2312.11805 (2023)."},{"key":"1875_CR62","doi-asserted-by":"crossref","unstructured":"Luo, Y. et al. An empirical study of catastrophic forgetting in large language models during continual fine-tuning. IEEE Transactions on Audio, Speech and Language Processing 33, 3776\u20133786 (2025).","DOI":"10.1109\/TASLPRO.2025.3606231"},{"key":"1875_CR63","unstructured":"Cui, G. et al. ULTRAFEEDBACK: Boosting language models with scaled AI feedback. In  Proceedings of the 41st International Conference on Machine Learning, vol. 235 of Proceedings of Machine Learning Research, (eds. Salakhutdinov, R. et al.) 9722\u20139744 (PMLR, 2024)."},{"key":"1875_CR64","unstructured":"Snorkel AI. Snorkel-mistral-pairrm-dpo-dataset https:\/\/huggingface.co\/datasets\/snorkelai\/Snorkel-Mistral-PairRM-DPO-Dataset (2024)."},{"key":"1875_CR65","doi-asserted-by":"publisher","first-page":"24824","DOI":"10.52202\/068431-1800","volume":"35","author":"J Wei","year":"2022","unstructured":"Wei, J. et al. Chain-of-thought prompting elicits reasoning in large language models. Advances in Neural Information Processing Systems 35, 24824\u201324837 (2022).","journal-title":"Advances in Neural Information Processing Systems"},{"key":"1875_CR66","doi-asserted-by":"crossref","unstructured":"Yun, J. et al. Med-PRM: medical reasoning models with stepwise, guideline-verified process rewards. In Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing, (eds. Christodoulopoulos, C., Chakraborty, T., Rose, C. & Peng, V.) 16565\u201316582 (Association for Computational Linguistics, 2025).","DOI":"10.18653\/v1\/2025.emnlp-main.837"},{"key":"1875_CR67","doi-asserted-by":"publisher","first-page":"1208","DOI":"10.1007\/s11524-025-01021-7","volume":"102","author":"FC Dos Santos","year":"2025","unstructured":"Dos Santos, F. C. et al. Global measures of HIV care accessibility across urban, suburban, and rural areas. Journal of Urban Health 102, 1208\u20131222 (2025).","journal-title":"Journal of Urban Health"},{"key":"1875_CR68","doi-asserted-by":"publisher","first-page":"509","DOI":"10.1513\/AnnalsATS.201307-246OT","volume":"10","author":"S Murthy","year":"2013","unstructured":"Murthy, S. & Adhikari, N. K. Global health care of the critically ill in low-resource settings. Annals of the American Thoracic Society 10, 509\u2013513 (2013).","journal-title":"Annals of the American Thoracic Society"},{"key":"1875_CR69","doi-asserted-by":"publisher","first-page":"565","DOI":"10.4081\/jphia.2016.565","volume":"7","author":"GIC Mbemba","year":"2016","unstructured":"Mbemba, G. I. C., Gagnon, M.-P. & Hamelin-Brabant, L. Factors influencing recruitment and retention of healthcare workers in rural and remote areas in developed and developing countries: an overview. Journal of Public Health in Africa 7, 565 (2016).","journal-title":"Journal of Public Health in Africa"},{"key":"1875_CR70","doi-asserted-by":"crossref","unstructured":"Ismail, A. I., Ibrahim, B. O., Adekanmbi, O. & Adebara, I. Retrieval-augmented generation meets local languages for improved drug information access and comprehension. In Proceedings of the Sixth Workshop on African Natural Language Processing (AfricaNLP 2025), (eds. Lignos, C., Abdulmumin, I. & Adelani, D.) 108\u2013114 (Association for Computational Linguistics, 2025).","DOI":"10.18653\/v1\/2025.africanlp-1.15"},{"key":"1875_CR71","unstructured":"Mutisya, F. et al. Retrieval-augmented clinical benchmarking for contextual model testing in Kenyan primary care: a methodology paper. Preprint at https:\/\/arxiv.org\/abs\/2507.14615 (2025)."},{"key":"1875_CR72","doi-asserted-by":"publisher","unstructured":"Duroux, D. Hivmedqa https:\/\/doi.org\/10.5281\/zenodo.19859260 (2025).","DOI":"10.5281\/zenodo.19859260"},{"key":"1875_CR73","doi-asserted-by":"publisher","unstructured":"Cardenal, G. Gonzalocardenalal\/medical_llm_evaluation: v1.0.1 https:\/\/doi.org\/10.5281\/zenodo.21498352 (2026).","DOI":"10.5281\/zenodo.21498352"}],"container-title":["Communications Medicine"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.nature.com\/articles\/s43856-026-01875-1","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/www.nature.com\/articles\/s43856-026-01875-1.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/www.nature.com\/articles\/s43856-026-01875-1.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,10,22]],"date-time":"2026-10-22T05:40:29Z","timestamp":1792647629000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.nature.com\/articles\/s43856-026-01875-1"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,8,29]]},"references-count":73,"journal-issue":{"issue":"1","published-online":{"date-parts":[[2026,12]]}},"alternative-id":["1875"],"URL":"https:\/\/doi.org\/10.1038\/s43856-026-01875-1","relation":{},"ISSN":["2730-664X"],"issn-type":[{"value":"2730-664X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,8,29]]},"assertion":[{"value":"15 August 2025","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"18 August 2026","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"29 August 2026","order":3,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"H.F.G. received unrestricted research grants from Gilead Sciences; fees for data and safety monitoring board membership from Merck; consulting fees from Merck; and grants from the Swiss National Science Foundation, the Gates Foundation (subcontractor), the Yvonne Jacob Foundation, Gilead Sciences, ViiV Healthcare. All the reported grants and remuneration were provided outside the context of this work. All other authors declare no competing interests.","order":1,"name":"Ethics","label":"Competing interests","group":{"name":"EthicsHeading","label":"Ethics"}}],"article-number":"552"}}