{"id":"W4417286768","doi":"10.64898/2025.12.11.693290","title":"Using GPT-4 to Automate the Generation of Lay Summaries for Cancer Publications","year":2025,"lang":"","type":"preprint","venue":"bioRxiv (Cold Spring Harbor Laboratory)","topic":"Text Readability and Simplification","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Ontario Institute for Cancer Research","funders":"University of Toronto","keywords":"Readability; Jargon; Reading (process); Sample (material); Data collection; Reading level; Comprehension","routes":{"ca_aff":true,"ca_fund":true,"ca_venue":false,"about_ca":true,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":["metaresearch"],"consensus_categories":[],"category_scores_codex":[0.01443489,0.002608244,0.001130924,0.008566787,0.000783154,0.004539748,0.001485127,0.00130001,0.02469622],"category_scores_gemma":[0.1055018,0.001046713,0.002527277,0.003333124,0.0005546615,0.003293657,0.004162278,0.001593867,0.01432766],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001449113,"about_ca_system_score_gemma":0.004115933,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001991037,"about_ca_topic_score_gemma":0.003465747,"domain_scores_codex":[0.9931716,0.003031955,0.001428431,0.001120782,0.001090245,0.0001569774],"domain_scores_gemma":[0.9197673,0.05521062,0.006637882,0.005575756,0.01138561,0.001422787],"domain_codex":null,"domain_gemma":"reporting","domain_candidate":"reporting","domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.001574095,0.0002181494,0.01029338,0.01588027,0.0006504955,0.001769091,0.007933342,0.008585637,0.01968836,0.00478607,0.2103278,0.7182934],"study_design_scores_gemma":[0.001447539,0.001740236,0.01932357,0.006256097,0.001443535,0.003522615,0.005448066,0.2417932,0.08886871,0.04600305,0.5833223,0.0008309619],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.02646496,0.001886165,0.5606327,0.00252119,0.001106886,0.00402971,0.04529101,0.351111,0.006956516],"genre_scores_gemma":[0.06523806,0.0009941628,0.8765675,0.000518676,0.0002426041,0.003392501,0.03747034,0.01066009,0.004916005],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9855651,"threshold_uncertainty_score":0.0826171,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0804196765853612,"score_gpt":0.3060255596561545,"score_spread":0.2256058830707933,"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."}}