Automating Evaluation of AI Text Generation in Healthcare with a Large Language Model (LLM)-as-a-Judge.

Saved in:
Bibliographic Details
Title: Automating Evaluation of AI Text Generation in Healthcare with a Large Language Model (LLM)-as-a-Judge.
Authors: Croxford E; Department of Biostatistics and Medical Informatics, University of Wisconsin, Madison, USA., Gao Y; Department of Biomedical Informatics, University of Colorado - Anschutz Medical, Aurora, USA., First E; Epic Systems, Verona, USA., Pellegrino N; Epic Systems, Verona, USA., Schnier M; Epic Systems, Verona, USA., Caskey J; Department of Medicine, University of Wisconsin, Madison, USA., Oguss M; Department of Medicine, University of Wisconsin, Madison, USA., Wills G; UW Health, Madison, USA., Chen G; Department of Biostatistics and Medical Informatics, University of Wisconsin, Madison, USA., Dligach D; Department of Medicine, University of Wisconsin, Madison, USA.; Department of Computer Science, Loyola University, Chicago, USA., Churpek MM; Department of Medicine, University of Wisconsin, Madison, USA.; Department of Biostatistics and Medical Informatics, University of Wisconsin, Madison, USA.; UW Health, Madison, USA., Mayampurath A; Department of Medicine, University of Wisconsin, Madison, USA.; Department of Biostatistics and Medical Informatics, University of Wisconsin, Madison, USA., Liao F; UW Health, Madison, USA.; BerbeeWalsh Department of Emergency Medicine, University of Wisconsin, Madison, USA., Goswami C; UW Health, Madison, USA., Wong KK; Epic Systems, Verona, USA.; UW Health, Madison, USA., Patterson BW; UW Health, Madison, USA.; BerbeeWalsh Department of Emergency Medicine, University of Wisconsin, Madison, USA., Afshar M; UW Health, Madison, USA.; Department of Medicine, University of Wisconsin, Madison, USA.
Source: MedRxiv : the preprint server for health sciences [medRxiv] 2025 May 06. Date of Electronic Publication: 2025 May 06.
Publication Type: Journal Article; Preprint
Journal Info: Country of Publication: United States NLM ID: 101767986 Publication Model: Electronic Cited Medium: Internet NLM ISO Abbreviation: medRxiv Subsets: PubMed not MEDLINE
Database: MEDLINE Ultimate
Description
DOI:10.1101/2025.04.22.25326219