{"id":"W4410347660","doi":"10.1101/2025.05.13.653828","title":"Large Language Models Can Extract Metadata for Annotation of Human Neuroimaging Publications","year":2025,"lang":"en","type":"preprint","venue":"bioRxiv (Cold Spring Harbor Laboratory)","topic":"Biomedical Text Mining and Ontologies","field":"Biochemistry, Genetics and Molecular Biology","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"National Institute on Drug Abuse","keywords":"Metadata; Annotation; Neuroimaging; Computer science; Natural language processing; Information retrieval; Data science; Artificial intelligence; World Wide Web; Psychology; Neuroscience","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":["metaresearch"],"consensus_categories":[],"category_scores_codex":[0.01530024,0.002090791,0.0006857021,0.006235689,0.001120907,0.004615332,0.001965943,0.002050656,0.006578154],"category_scores_gemma":[0.07202812,0.0008838064,0.001799107,0.003604305,0.0009187221,0.006830748,0.004426381,0.002442758,0.01130334],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00179709,"about_ca_system_score_gemma":0.003997877,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005057836,"about_ca_topic_score_gemma":0.01777928,"domain_scores_codex":[0.9898453,0.005281405,0.0009911492,0.001617267,0.0019648,0.0003001233],"domain_scores_gemma":[0.9472686,0.03394466,0.002958992,0.009347568,0.005742274,0.0007378759],"domain_codex":null,"domain_gemma":"methods","domain_candidate":"methods","domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.001455919,0.0005667753,0.02242414,0.003942287,0.0009094097,0.0007269288,0.00309894,0.03736451,0.03081167,0.02610504,0.1892518,0.6833426],"study_design_scores_gemma":[0.0002130869,0.0003878915,0.01417111,0.0009012769,0.000436521,0.000719277,0.001630776,0.6073826,0.05238222,0.1072255,0.2141603,0.0003894242],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.04760449,0.002559629,0.784364,0.004715575,0.0008869548,0.0007407008,0.02836052,0.1144931,0.01627508],"genre_scores_gemma":[0.2656749,0.001097911,0.6517713,0.001512853,0.0003550524,0.001117773,0.06465247,0.006334084,0.007483598],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9846998,"threshold_uncertainty_score":0.0809164,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03010373277821184,"score_gpt":0.2940016720894056,"score_spread":0.2638979393111938,"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."}}