{"id":"W4411884041","doi":"10.1038/s41467-025-60762-w","title":"Extracting circumstances of Covid-19 transmission from free text with large language models","year":2025,"lang":"en","type":"article","venue":"Nature Communications","topic":"Topic Modeling","field":"Computer Science","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"Jewish General Hospital","funders":"Institut Pasteur; Agence Nationale de la Recherche","keywords":"Coronavirus disease 2019 (COVID-19); Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2); 2019-20 coronavirus outbreak; Transmission (telecommunications); Computer science; Pandemic; Virology; Computational biology; Medicine; Biology; Outbreak; Telecommunications; Disease; Infectious disease (medical specialty)","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":[0.004774268,0.001562456,0.0006042191,0.003767987,0.0005199339,0.002204181,0.001001206,0.001449695,0.001942167],"category_scores_gemma":[0.01984868,0.0004633547,0.001608973,0.00148633,0.0004692676,0.003906059,0.001889039,0.001816998,0.002431824],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0008719508,"about_ca_system_score_gemma":0.001081915,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00379956,"about_ca_topic_score_gemma":0.00604802,"domain_scores_codex":[0.9956805,0.002636066,0.0004319345,0.0008082801,0.0002883051,0.0001549734],"domain_scores_gemma":[0.9758976,0.02036219,0.001520742,0.0009479605,0.0009575253,0.0003139477],"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":[0.002235654,0.001205729,0.2470668,0.004835603,0.001076436,0.004097173,0.011123,0.09143491,0.02843626,0.01036413,0.06419874,0.5339255],"study_design_scores_gemma":[0.0001554982,0.0004708997,0.05763054,0.0006907017,0.0003521572,0.001054168,0.005655902,0.8559945,0.008225872,0.02974748,0.03975514,0.0002671657],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.5443839,0.003399444,0.3685994,0.005470646,0.0009216347,0.001338152,0.06124216,0.008114026,0.006530581],"genre_scores_gemma":[0.8129944,0.0006785062,0.1336938,0.0006713356,0.0004626388,0.0008705215,0.04848485,0.0002062158,0.001937629],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.004774268,"threshold_uncertainty_score":0.02524906,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0279004093832934,"score_gpt":0.3268215091203649,"score_spread":0.2989210997370715,"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."}}