{"id":"W7115714437","doi":"10.48448/s1j9-wp94","title":"Domain-specific Pretrained Encoder Transformers for the Identification of Methodologically Rigorous Systematic Reviews: A Retrospective Modeling Study","year":2025,"lang":"","type":"other","venue":"Underline Science Inc.","topic":"","field":"","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"McMaster University","funders":"","keywords":"Systematic review; Transformer; Identification (biology); Data extraction; Encoder; Narrative; Ground truth","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":["metaresearch"],"consensus_categories":["metaresearch"],"category_scores_codex":[0.3139907,0.002477747,0.007482216,0.01601969,0.0019141,0.007089776,0.003734171,0.002522653,0.01689694],"category_scores_gemma":[0.6857257,0.002895986,0.01740144,0.01710149,0.002298669,0.009146824,0.006544685,0.002730629,0.002377194],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.007541578,"about_ca_system_score_gemma":0.02882284,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004065169,"about_ca_topic_score_gemma":0.009369295,"domain_scores_codex":[0.5753801,0.2491135,0.1309275,0.01755727,0.02388869,0.0031329],"domain_scores_gemma":[0.1926054,0.6572848,0.08567587,0.02927487,0.0334839,0.001675118],"domain_codex":null,"domain_gemma":"evaluation","domain_candidate":"evaluation","domain_consensus":null,"study_design_codex":"systematic_review","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.006871613,0.0006148089,0.1635485,0.4582647,0.08281711,0.001094615,0.01305717,0.007204039,0.001400128,0.01940631,0.02183869,0.2238824],"study_design_scores_gemma":[0.009142818,0.005346861,0.1427042,0.372142,0.2541349,0.003026799,0.01291401,0.05323027,0.006187086,0.0416179,0.09793287,0.001620404],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1661488,0.1504068,0.4082671,0.01682412,0.002168987,0.1982065,0.04557481,0.002607326,0.009795556],"genre_scores_gemma":[0.4596546,0.01549729,0.2773883,0.003947112,0.0003488894,0.232805,0.008513629,0.0004051917,0.001440065],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.6860093,"threshold_uncertainty_score":0.8459716,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1202096400569525,"score_gpt":0.3808638434936585,"score_spread":0.260654203436706,"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."}}