Systematic benchmarking demonstrates large language models have not reached the diagnostic accuracy of traditional rare-disease decision support tools.

Saved in:
Bibliographic Details
Title: Systematic benchmarking demonstrates large language models have not reached the diagnostic accuracy of traditional rare-disease decision support tools.
Authors: Reese JT; Division of Environmental Genomics and Systems Biology, Lawrence Berkeley National Laboratory, Berkeley, CA, USA., Chimirri L; Berlin Institute of Health at Charité Universitätsmedizin Berlin, Berlin, Germany., Bridges Y; William Harvey Research Institute, Barts and The London School of Medicine and Dentistry, Queen Mary University of London, London, UK., Danis D; Berlin Institute of Health at Charité Universitätsmedizin Berlin, Berlin, Germany., Caufield JH; Division of Environmental Genomics and Systems Biology, Lawrence Berkeley National Laboratory, Berkeley, CA, USA., Gargano MA; The Jackson Laboratory for Genomic Medicine, 10 Discovery Drive, Farmington, CT, USA., Kroll C; William Harvey Research Institute, Barts and The London School of Medicine and Dentistry, Queen Mary University of London, London, UK., Schmeder A; ScienceIT, Lawrence Berkeley National Laboratory, Berkeley, CA, USA., Liu F; ScienceIT, Lawrence Berkeley National Laboratory, Berkeley, CA, USA., Wissink K; Berlin Institute of Health at Charité Universitätsmedizin Berlin, Berlin, Germany., McMurry JA; University of North Carolina at Chapel Hill, Chapel Hill, NC, USA., Graefe ASL; Berlin Institute of Health at Charité Universitätsmedizin Berlin, Berlin, Germany., Niyonkuru E; The Jackson Laboratory for Genomic Medicine, 10 Discovery Drive, Farmington, CT, USA., Korn DR; University of North Carolina at Chapel Hill, Chapel Hill, NC, USA., Casiraghi E; Division of Environmental Genomics and Systems Biology, Lawrence Berkeley National Laboratory, Berkeley, CA, USA.; AnacletoLab, Dipartimento di Informatica, Università degli Studi di Milano, Milano, Italy.; ELLIS-European Laboratory for Learning and Intelligent Systems, Milan UnitMilan Unit, Milan, Italy., Valentini G; AnacletoLab, Dipartimento di Informatica, Università degli Studi di Milano, Milano, Italy.; ELLIS-European Laboratory for Learning and Intelligent Systems, Milan UnitMilan Unit, Milan, Italy., Jacobsen JOB; William Harvey Research Institute, Barts and The London School of Medicine and Dentistry, Queen Mary University of London, London, UK., Haendel M; University of North Carolina at Chapel Hill, Chapel Hill, NC, USA., Smedley D; William Harvey Research Institute, Barts and The London School of Medicine and Dentistry, Queen Mary University of London, London, UK., Mungall CJ; Division of Environmental Genomics and Systems Biology, Lawrence Berkeley National Laboratory, Berkeley, CA, USA., Robinson PN; Berlin Institute of Health at Charité Universitätsmedizin Berlin, Berlin, Germany. peter.robinson@bih-charite.de.; The Jackson Laboratory for Genomic Medicine, 10 Discovery Drive, Farmington, CT, USA. peter.robinson@bih-charite.de.; ELLIS-European Laboratory for Learning and Intelligent Systems, Milan UnitMilan Unit, Milan, Italy. peter.robinson@bih-charite.de.
Source: European journal of human genetics : EJHG [Eur J Hum Genet] 2026 Apr; Vol. 34 (4), pp. 498-504. Date of Electronic Publication: 2026 Feb 24.
Publication Type: Journal Article
Journal Info: Publisher: Nature Publishing Group Country of Publication: England NLM ID: 9302235 Publication Model: Print-Electronic Cited Medium: Internet ISSN: 1476-5438 (Electronic) Linking ISSN: 10184813 NLM ISO Abbreviation: Eur J Hum Genet Subsets: MEDLINE
Database: MEDLINE Ultimate
Description
ISSN:1476-5438
DOI:10.1038/s41431-026-02054-5