{"id":"W4409416357","doi":"10.2196/62909","title":"Summarizing Online Patient Conversations Using Generative Language Models: Experimental and Comparative Study","year":2025,"lang":"en","type":"article","venue":"JMIR Medical Informatics","topic":"Machine Learning in Healthcare","field":"Computer Science","cited_by":4,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Computer science; Language model; Task (project management); Natural language processing; Generative grammar; Artificial intelligence; Information retrieval; Transformer; Data science","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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0002017939,0.0001376564,0.0002242506,0.000139163,0.000219124,0.0001035567,0.0003264607,0.00007195578,0.00001517605],"category_scores_gemma":[0.00004966011,0.0001183616,0.00002413873,0.0002943634,0.00008421001,0.0005180605,0.000516068,0.0003677649,0.000002809314],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001047198,"about_ca_system_score_gemma":0.0002218067,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000151254,"about_ca_topic_score_gemma":0.00003691067,"domain_scores_codex":[0.9985199,0.000146407,0.0005006377,0.0001378398,0.0004980307,0.0001971672],"domain_scores_gemma":[0.9991981,0.0001590555,0.000135697,0.0002659123,0.00008186148,0.0001594198],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"qualitative","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00001588,0.0009437751,0.004674103,0.0001537273,0.0001101041,0.00002864355,0.933435,0.004248652,0.00006248985,0.03677566,0.0006331686,0.01891879],"study_design_scores_gemma":[0.000466488,0.0001511025,0.0002893968,0.00006973173,0.000004478657,0.00000573125,0.08802561,0.9105834,0.0001031336,0.00008458815,0.0001165308,0.00009981132],"study_design_candidate":"qualitative","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8517567,0.0001627175,0.1459014,0.0003463068,0.0001696819,0.0005571297,0.000005970563,0.00009650746,0.001003578],"genre_scores_gemma":[0.9513428,0.000002512287,0.04721083,0.00134784,0.00002364285,0.00003353283,0.0000138314,0.00000289002,0.00002217461],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.9063348,"threshold_uncertainty_score":0.4826645,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0673433530193716,"score_gpt":0.3960594878841782,"score_spread":0.3287161348648066,"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."}}