{"id":"W4400343221","doi":"10.48550/arxiv.2407.00541","title":"Answering real-world clinical questions using large language model based systems","year":2024,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Topic Modeling","field":"Computer Science","cited_by":5,"is_retracted":false,"has_abstract":true,"ca_institutions":"Hospital for Sick Children","funders":"","keywords":"Computer science; Question answering; Natural language processing","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0206381,0.001569143,0.001295518,0.004624103,0.0009593707,0.00451987,0.002036396,0.002657248,0.003355204],"category_scores_gemma":[0.06444328,0.0007162907,0.00215818,0.002316381,0.0008980848,0.003491808,0.002938768,0.00181502,0.001526967],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002080587,"about_ca_system_score_gemma":0.002806833,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003563901,"about_ca_topic_score_gemma":0.00745608,"domain_scores_codex":[0.9853123,0.01029756,0.001456846,0.001603178,0.001125512,0.000204582],"domain_scores_gemma":[0.9033003,0.08555654,0.003513852,0.003861625,0.003165489,0.0006020534],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.001598095,0.001121857,0.01701904,0.004990195,0.001898234,0.001502951,0.005693789,0.1718238,0.02945524,0.0152219,0.02730969,0.7223652],"study_design_scores_gemma":[0.000535529,0.0005743901,0.003667235,0.0005608612,0.0007196688,0.0005153119,0.001533393,0.8826836,0.01819226,0.0615567,0.02918578,0.0002752417],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.05321427,0.002655005,0.9141831,0.005394654,0.0001867977,0.001871651,0.003729665,0.01637257,0.002392329],"genre_scores_gemma":[0.2545375,0.0008539163,0.7355561,0.001579635,0.0002342785,0.001193643,0.004773736,0.0002831909,0.0009879884],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.0206381,"threshold_uncertainty_score":0.1091461,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1487119472503954,"score_gpt":0.2780776775516317,"score_spread":0.1293657303012362,"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."}}