Evaluating the effectiveness of biomedical fine-tuning for large language models on clinical tasks
Fine-tuning LLMs on biomedical data may not yield the anticipated benefits. Alternative approaches, such as retrieval augmentation, should be further explored for effective and reliable clinical integration of LLMs.
Authors
- Lisa C. Adams (ORCID: https://orcid.org/0000-0001-5836-4542)
- Keno K. Bressem (ORCID: https://orcid.org/0000-0001-9249-8624)
- Amin Dada (ORCID: https://orcid.org/0000-0003-4016-7799)
- Madhumita Sushil (ORCID: https://orcid.org/0000-0001-7884-0526)
- Daniel Truhn (ORCID: https://orcid.org/0000-0002-9605-0728)
- Jens Kleesiek (ORCID: https://orcid.org/0000-0001-8686-0682)
- Tianyu Han (ORCID: https://orcid.org/0000-0002-0792-9272)
- Felix Busch (ORCID: https://orcid.org/0000-0001-9770-8555)
- Marcus R. Makowski (ORCID: https://orcid.org/0000-0001-8778-647X)
- Felix J. Dorfner (ORCID: https://orcid.org/0009-0003-5914-3233)
Institutions
- Harvard University (US)
- German Cancer Research Center (DE)
- University of California, San Francisco (US)
- TU Dortmund University (DE)
- TUM Klinikum (DE)
- Humboldt-Universität zu Berlin (DE)
- Massachusetts General Hospital (US)
- Essen University Hospital (DE)
- Universitätsklinikum Aachen (DE)
- Athinoula A. Martinos Center for Biomedical Imaging (US)
- Cancer Research Center (US)
- Deutsches Herzzentrum München (DE)
- Deutsches Konsortium für Translationale Krebsforschung (DE)
- Freie Universität Berlin (DE)
- Charité - Universitätsmedizin Berlin (DE)
Publication Details
- Journal
- Journal of the American Medical Informatics Association
- Published
- 2025-04-07
- DOI
- https://doi.org/10.1093/jamia/ocaf045
- Citations
- 23
- Primary Topic
- Artificial Intelligence in Healthcare and Education
- Type
- article
- Field-Weighted Citation Impact
- 15.05
Funders
- Institute for Advanced Study
- Bundesministerium für Bildung und Forschung
- Technische Universität München