ResearchBenchmarksMedical Evaluation on MediEvalFollow49.3AccuracyGPT-4o45.55646.52847.548.472Dec 23, 2025Evaluation ResultsMethodMethodLinksAccuracyPrecisionRecallF1 ScoreF1 Score (Q1)F1 Score (Q2)F1 Score (Q3)F1 Score (Q4)HSRTIRGPT-4oModel=GPT-4o, Adaptati...Model=GPT-4o, Adaptation=Base generative form2025.1249.331.65029.166.3500046.7—GPT-5Model=GPT-5, Adaptatio...Model=GPT-5, Adaptation=Base generative form2025.1245.736.446.730.468.353.30019.3—