{"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":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow","scholarly_communication"],"consensus_categories":[],"category_scores_codex":[0.002062153,0.0005903923,0.0006910331,0.0004817425,0.001029357,0.00108719,0.002410699,0.0005011113,0.00002664171],"category_scores_gemma":[0.0009402462,0.0005533407,0.0002872342,0.002261118,0.0002847869,0.0006472802,0.001019136,0.0004251713,0.000008939751],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0006403835,"about_ca_system_score_gemma":0.003777696,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0006048318,"about_ca_topic_score_gemma":0.0001100028,"domain_scores_codex":[0.9954469,0.0003474716,0.001353887,0.001611602,0.000555571,0.0006845673],"domain_scores_gemma":[0.9917728,0.0003919016,0.001017072,0.003409752,0.003158103,0.0002503968],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00003575253,0.0003572961,0.004228519,0.001033129,0.0003319312,4.212889e-7,0.0003503539,0.02647007,0.7880363,0.1778131,0.00115707,0.0001860348],"study_design_scores_gemma":[0.0003339218,0.00006219865,0.0586484,0.0005024645,0.0002931014,8.428402e-9,0.000008303246,0.454423,0.414293,0.00002771141,0.07052974,0.0008781645],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2169514,0.0009374967,0.7640941,0.008927049,0.00305029,0.003936629,0.001746178,0.0003464835,0.00001048105],"genre_scores_gemma":[0.9131439,0.0002039854,0.08362123,0.000525324,0.0005716121,0.001836665,0.000001284098,0.00004779897,0.00004817978],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.6961926,"threshold_uncertainty_score":0.9999498,"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."}}