{"id":"W4405597099","doi":"10.1038/s41746-024-01366-4","title":"Probabilistic medical predictions of large language models","year":2024,"lang":"en","type":"article","venue":"npj Digital Medicine","topic":"Topic Modeling","field":"Computer Science","cited_by":34,"is_retracted":false,"has_abstract":true,"ca_institutions":"Artificial Intelligence in Medicine (Canada)","funders":"National Institutes of Health","keywords":"Probabilistic logic; Computer science; Artificial intelligence","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":[0.0003086553,0.000102148,0.0001809717,0.0001222373,0.00002600351,0.00005010047,0.000500673,0.0000667491,0.00008837994],"category_scores_gemma":[0.0004342242,0.00007358271,0.00004489931,0.0003469933,0.00009407267,0.0005969558,0.0001880584,0.0001743797,0.00002428624],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00003053424,"about_ca_system_score_gemma":0.0001358401,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001621311,"about_ca_topic_score_gemma":0.000005166288,"domain_scores_codex":[0.9984635,0.00001449358,0.0003202724,0.0003058653,0.0006841146,0.0002117721],"domain_scores_gemma":[0.9991649,0.0001797777,0.00002834857,0.0004158086,0.00005003119,0.0001611733],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00000323623,0.000119465,0.00004791218,0.000297968,0.00004648744,0.0002089137,0.005575152,0.0008564747,0.00008890382,0.8913528,0.004084505,0.09731824],"study_design_scores_gemma":[0.0002351595,0.0001018631,0.00003127433,0.000473412,0.000009871455,0.00006083362,0.0001250383,0.9677889,0.00002200688,0.02859833,0.002477292,0.00007605473],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01290781,0.001319106,0.9421204,0.003946608,0.0007406425,0.0001359348,0.00001946993,0.0003952871,0.03841471],"genre_scores_gemma":[0.9979385,0.00001271947,0.0009106128,0.0001719468,0.0002967748,0.00001151552,0.00001015336,0.000009253104,0.0006385271],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9850307,"threshold_uncertainty_score":0.3000616,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01921885941239498,"score_gpt":0.2821147625464776,"score_spread":0.2628959031340826,"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."}}