{"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":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0008264777,0.0003270507,0.0004466467,0.0004761881,0.0001602667,0.0002826304,0.001504124,0.0003149678,0.000006551089],"category_scores_gemma":[0.00003505719,0.0003863932,0.0003175171,0.0006076206,0.00005246235,0.0002288359,0.002757768,0.001119797,0.0000518755],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0003947399,"about_ca_system_score_gemma":0.0006198362,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001446333,"about_ca_topic_score_gemma":0.0003573999,"domain_scores_codex":[0.9972882,0.0002324927,0.0004389419,0.001448508,0.0001366082,0.0004552971],"domain_scores_gemma":[0.9975345,0.0001299741,0.0002138033,0.001785654,0.0001162404,0.0002198119],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.000003424884,0.00004180126,0.0007855935,0.0001504843,0.00003968038,0.0003587599,0.0001294734,0.8128902,0.00002234482,0.1854504,0.00003440683,0.00009341365],"study_design_scores_gemma":[0.0002243558,0.000008857445,0.00005614108,0.0004782916,0.00009580491,0.000002346503,0.00005653281,0.9909021,0.00001010718,0.007713902,0.00007059532,0.0003809283],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2072747,0.0001373007,0.7880199,0.00004794258,0.00150603,0.0002371218,0.00003387423,0.0006000997,0.002143032],"genre_scores_gemma":[0.9783337,0.00004478889,0.01909214,0.00006758601,0.0002301537,0.000001186167,0.00001009224,0.00003377437,0.00218664],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.7710589,"threshold_uncertainty_score":0.9998588,"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."}}