{"id":"W7077890825","doi":"10.48448/gamm-6t13","title":"Zero-Shot ATC Coding with Large Language Models for Clinical Assessments","year":2025,"lang":"en","type":"other","venue":"Underline Science Inc.","topic":"Geochemistry and Geologic Mapping","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"","keywords":"Coding (social sciences); Bottleneck; Health care; Ontology; Process (computing); Data integration","routes":{"ca_aff":true,"ca_fund":false,"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":[],"consensus_categories":[],"category_scores_codex":[0.002727271,0.001304607,0.0007470332,0.001839569,0.0008810935,0.002685911,0.002030412,0.001372207,0.005339147],"category_scores_gemma":[0.02438551,0.000723858,0.001864596,0.001357416,0.0008589117,0.003575098,0.003305511,0.003419996,0.004172201],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002520513,"about_ca_system_score_gemma":0.006123495,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.03465842,"about_ca_topic_score_gemma":0.06355732,"domain_scores_codex":[0.9965717,0.001488327,0.0003076966,0.0007257372,0.0007339744,0.0001726571],"domain_scores_gemma":[0.9876792,0.008369737,0.0005750188,0.001808242,0.001272369,0.0002954119],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"not_applicable","study_design_scores_codex":[0.001034668,0.0004730077,0.01534674,0.0008659049,0.0002834858,0.00108271,0.001979631,0.1423446,0.01496898,0.01975815,0.0657105,0.7361516],"study_design_scores_gemma":[0.00006495786,0.00007185491,0.001358696,0.00009831288,0.0000635277,0.0002824907,0.0003997122,0.9365627,0.006639646,0.03499358,0.01939208,0.0000724978],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.03514439,0.0009353834,0.8942055,0.002478408,0.0002686386,0.000502691,0.009319896,0.05319455,0.003950526],"genre_scores_gemma":[0.3613822,0.0004342594,0.6152133,0.001403217,0.0001228361,0.0005266402,0.01630989,0.001593862,0.003013839],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.03465842,"threshold_uncertainty_score":0.0689134,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.06960542250939906,"score_gpt":0.3862814749890439,"score_spread":0.3166760524796448,"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."}}