{"id":"W4414132264","doi":"10.1017/rsm.2025.10031","title":"StudyTypeTeller—Large language models to automatically classify research study types for systematic reviews","year":2025,"lang":"en","type":"article","venue":"Research Synthesis Methods","topic":"Topic Modeling","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"McMaster University","funders":"Universität Zürich; Schweizerischer Nationalfonds zur Förderung der Wissenschaftlichen Forschung","keywords":"Generative grammar; Transformer; Systematic review; Language model; Scientific literature; Encoder","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":"codex-gemma-dda1882f352a","candidate_categories":["metaresearch"],"consensus_categories":["metaresearch"],"category_scores_codex":[0.128479,0.0002676705,0.001163876,0.001490293,0.0007509533,0.0005946782,0.003614338,0.0001497793,0.00002787248],"category_scores_gemma":[0.050952,0.0002023419,0.0001997918,0.002938013,0.0000757496,0.0003279462,0.002077681,0.0007182938,0.0001780307],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0003652933,"about_ca_system_score_gemma":0.0004450747,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00007092837,"about_ca_topic_score_gemma":0.00004687865,"domain_scores_codex":[0.9675128,0.02666615,0.001185495,0.00127153,0.001852188,0.001511769],"domain_scores_gemma":[0.9570294,0.0384276,0.00009332129,0.00303102,0.001100119,0.0003185661],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0001218837,0.002832184,0.00007371175,0.04303597,0.0007245846,0.00006283352,0.03381414,0.0008599829,0.006832332,0.525389,0.01243667,0.3738167],"study_design_scores_gemma":[0.0004020437,0.0004075276,0.00007221654,0.006535455,0.00005300429,0.000002168342,0.007373511,0.9498599,0.003162141,0.02844194,0.003316112,0.0003740476],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.001639106,0.00168452,0.9781677,0.001845038,0.0001668827,0.01100889,0.00000345825,0.0001959947,0.005288409],"genre_scores_gemma":[0.1196645,0.00005090041,0.8674881,0.0001189145,0.00009359236,0.008151558,3.594257e-7,0.00003506635,0.004396981],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.9489998,"threshold_uncertainty_score":0.9570422,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.4744790487850936,"score_gpt":0.5983134327456537,"score_spread":0.1238343839605601,"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."}}