{"id":"W4412629089","doi":"10.2196/66476","title":"Improving Large Language Models’ Summarization Accuracy by Adding Highlights to Discharge Notes: Comparative Evaluation","year":2025,"lang":"en","type":"article","venue":"JMIR Medical Informatics","topic":"Machine Learning in Healthcare","field":"Computer Science","cited_by":13,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Automatic summarization; Terminology; Computer science; Process (computing); Interface (matter); Artificial intelligence; Test (biology); Information retrieval; Natural language processing; Data science; Linguistics; Programming language","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":true,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.02377836,0.002629824,0.001773746,0.003122201,0.0009959082,0.002745941,0.002539992,0.002097916,0.003599312],"category_scores_gemma":[0.0901353,0.000512645,0.002296771,0.001913166,0.0007707664,0.004279885,0.002395352,0.002584892,0.001880045],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002008332,"about_ca_system_score_gemma":0.001980374,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.008441411,"about_ca_topic_score_gemma":0.008601536,"domain_scores_codex":[0.9831094,0.01153355,0.001960405,0.00190769,0.001233683,0.0002551689],"domain_scores_gemma":[0.8731196,0.1106649,0.002572012,0.005193685,0.007308668,0.001141175],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.01185785,0.003508315,0.02373931,0.007132765,0.003662415,0.0006612107,0.002560733,0.1490497,0.006359547,0.001347549,0.03772976,0.7523909],"study_design_scores_gemma":[0.002706111,0.006449518,0.02104354,0.0008631228,0.003242429,0.0005796725,0.002193749,0.9264153,0.0140922,0.004812648,0.01730321,0.0002984325],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.7580062,0.02927909,0.1523493,0.004063105,0.002395151,0.002924462,0.01551642,0.02543809,0.01002819],"genre_scores_gemma":[0.786999,0.004002949,0.1634375,0.0009518229,0.0006748151,0.001173706,0.03910756,0.001027469,0.002625168],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.02377836,"threshold_uncertainty_score":0.1257535,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02851150607305516,"score_gpt":0.375509264051678,"score_spread":0.3469977579786229,"validation_status":"score_only:v0-immature-baseline","note":"Baseline scores from an immature model (maturity gate not passed). Scores rank; they never assert a category."}}