{"id":"W4414155525","doi":"10.1002/cesm.70045","title":"Artificial Intelligence Search Tools for Evidence Synthesis: Comparative Analysis and Implementation Recommendations","year":2025,"lang":"en","type":"article","venue":"Cochrane Evidence Synthesis and Methods","topic":"Artificial Intelligence in Healthcare and Education","field":"Medicine","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"Public Health Agency of Canada; Canadian Agency for Drugs and Technologies in Health","funders":"","keywords":"Leverage (statistics); Agency (philosophy); Automation; Information access; Information system; Focus (optics)","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":"codex-gemma-dda1882f352a","candidate_categories":["metaresearch"],"consensus_categories":[],"category_scores_codex":[0.005412658,0.0002184291,0.0007107581,0.0005885803,0.0004990047,0.0002236623,0.0001375839,0.0001290619,0.0002958482],"category_scores_gemma":[0.01396483,0.000199245,0.0001645143,0.001192893,0.0002320043,0.0006508419,0.00006294484,0.000189648,0.000005916264],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001272134,"about_ca_system_score_gemma":0.0003668953,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0006601877,"about_ca_topic_score_gemma":0.0003328595,"domain_scores_codex":[0.9970657,0.00084687,0.0008771545,0.0006320071,0.0002202753,0.0003580451],"domain_scores_gemma":[0.962705,0.03594598,0.0001678221,0.0003864519,0.0006033774,0.0001913676],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.0002792617,0.00006117152,0.004526575,0.0005615056,0.0003017399,6.052233e-7,0.002152821,0.00001376288,0.007955967,0.003105043,0.0001507043,0.9808909],"study_design_scores_gemma":[0.00001888307,0.0002567183,0.01893991,0.003447903,0.004104836,0.000008331002,0.03022129,0.009775752,0.9267235,0.005565763,0.0006206764,0.000316392],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1736081,0.01027412,0.7810093,0.03281509,0.000340198,0.001738084,0.00002420169,0.00005928733,0.0001316106],"genre_scores_gemma":[0.8448301,0.0107325,0.1430098,0.0003665292,0.0001232058,0.0008406087,0.00001003362,0.000008887747,0.00007840529],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9805744,"threshold_uncertainty_score":0.994341,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.4332274836753609,"score_gpt":0.6187690955254639,"score_spread":0.185541611850103,"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."}}