{"id":"W4407274503","doi":"10.1093/jncics/pkaf021","title":"The use of large language models to enhance cancer clinical trial educational materials","year":2025,"lang":"en","type":"article","venue":"JNCI Cancer Spectrum","topic":"Artificial Intelligence in Healthcare and Education","field":"Medicine","cited_by":9,"is_retracted":false,"has_abstract":true,"ca_institutions":"Artificial Intelligence in Medicine (Canada)","funders":"National Cancer Institute","keywords":"Readability; Automatic summarization; Clinical trial; Comprehension; Medicine; Plain language; Protocol (science); MEDLINE; Medical education; Alternative medicine; Computer science; Artificial intelligence; Pathology","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0005122267,0.0001041847,0.0002834237,0.0000716444,0.0001385631,0.00003560317,0.000132548,0.00009479948,0.0007308024],"category_scores_gemma":[0.0003657135,0.00007614128,0.00008384052,0.0002897283,0.00006273529,0.00009093564,0.00004002762,0.0001573933,0.00002681891],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002347094,"about_ca_system_score_gemma":0.00173613,"about_ca_topic_candidate":true,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0124489,"about_ca_topic_score_gemma":0.006524387,"domain_scores_codex":[0.998429,0.00008612394,0.0007216447,0.0002606483,0.0002062459,0.0002963546],"domain_scores_gemma":[0.9987437,0.0004416706,0.000157712,0.0003837993,0.0001677109,0.0001054502],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.07958127,0.002508634,0.03173523,0.0006815903,0.0007207559,0.000007155891,0.01309463,0.0007061097,0.01327223,0.0734328,0.4419158,0.3423438],"study_design_scores_gemma":[0.003233287,0.001252213,0.02613379,0.001485485,0.0003936871,0.000003307186,0.002739402,0.001216117,0.7338617,0.01662936,0.2124781,0.0005734464],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9272198,0.001203638,0.0004839109,0.06311099,0.006688337,0.0008916698,0.00007159262,0.00001869607,0.0003113495],"genre_scores_gemma":[0.9819398,0.001513989,0.0001955611,0.004021326,0.002536449,0.0004450101,0.00001344316,0.00001284152,0.009321615],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.7205895,"threshold_uncertainty_score":0.9941273,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2787258091545623,"score_gpt":0.5640962906693894,"score_spread":0.2853704815148271,"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."}}