{"id":"W7130731173","doi":"10.1109/swc65939.2025.00239","title":"Small Language Models for Emergency Departments Decision Support: A Benchmark Study","year":2025,"lang":"","type":"article","venue":"","topic":"Machine Learning in Healthcare","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Rockyview General Hospital; University of Calgary","funders":"","keywords":"Benchmark (surveying); Workflow; Variety (cybernetics); Key (lock); Language model; Focus (optics); Clinical decision making","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":[0.005256929,0.001624238,0.0008271699,0.001417837,0.0005293859,0.001371596,0.001949349,0.001490083,0.00238494],"category_scores_gemma":[0.02031325,0.000400394,0.0008634935,0.00164266,0.000714779,0.001868282,0.001174339,0.00196325,0.0009616036],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002091429,"about_ca_system_score_gemma":0.002101372,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01780257,"about_ca_topic_score_gemma":0.02610961,"domain_scores_codex":[0.9970636,0.001694827,0.0002454466,0.0005369383,0.0003038238,0.0001554221],"domain_scores_gemma":[0.9854917,0.01144101,0.0004084129,0.001119435,0.001111695,0.0004277935],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"observational","study_design_scores_codex":[0.002421007,0.002165838,0.015959,0.002295101,0.0008048429,0.0006330727,0.0003740756,0.7012414,0.003915948,0.004152313,0.05995443,0.2060829],"study_design_scores_gemma":[0.0007132689,0.0008461651,0.005002924,0.0001500474,0.0001402867,0.0002438447,0.0003228966,0.9695052,0.00491606,0.006849315,0.0112474,0.00006254737],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8459229,0.0182878,0.07969644,0.006371445,0.0007415194,0.001167155,0.02466752,0.008972531,0.01417273],"genre_scores_gemma":[0.8710012,0.002696589,0.08743757,0.00117983,0.0002640059,0.0005066582,0.03359466,0.0003580308,0.00296126],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.01780257,"threshold_uncertainty_score":0.03539789,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04067461427722333,"score_gpt":0.3637823539918068,"score_spread":0.3231077397145834,"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."}}