{"id":"W4409187149","doi":"10.1007/s10940-025-09606-w","title":"Beyond Traditional Risk Scores: Tackling LS/CMI Offender Misclassifications with Machine Learning","year":2025,"lang":"en","type":"article","venue":"Journal of Quantitative Criminology","topic":"Artificial Intelligence in Law","field":"Social Sciences","cited_by":2,"is_retracted":false,"has_abstract":false,"ca_institutions":"Université Laval; Ministère des Transports; Université de Montréal","funders":"","keywords":"Artificial intelligence; Machine learning; Computer science; Psychology","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":[],"consensus_categories":[],"category_scores_codex":[0.001110869,0.0001335696,0.0002982738,0.0003998683,0.0009597025,0.00006761118,0.0003554245,0.0001289678,0.0003331382],"category_scores_gemma":[0.002889817,0.0001107309,0.0001192883,0.0003504886,0.001462459,0.0003534682,0.00002243041,0.0007721331,0.00003201782],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001558432,"about_ca_system_score_gemma":0.0006895685,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000456279,"about_ca_topic_score_gemma":0.001596728,"domain_scores_codex":[0.9977855,0.000766336,0.0005739335,0.0001927376,0.0003705748,0.000310905],"domain_scores_gemma":[0.9962419,0.001985993,0.0006958859,0.0001049601,0.0008723355,0.00009892503],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.0001735308,0.0001480019,0.01431049,0.00001031007,0.0001869123,0.00002440289,0.01271677,0.001135674,0.0007449709,0.9665298,0.0009811327,0.003038021],"study_design_scores_gemma":[0.001244872,0.002760605,0.03942873,0.0004878867,0.0008927482,0.00008982552,0.1450643,0.004335715,0.005783522,0.7210988,0.07805441,0.000758617],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":"theoretical_or_conceptual","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.7925238,0.001943914,0.1369909,0.01330507,0.001514403,0.0002779462,0.00002999704,0.00005753398,0.05335636],"genre_scores_gemma":[0.986311,0.0004709959,0.01201638,0.0002620986,0.0001681362,0.00000572068,0.000004610579,0.00001126823,0.0007498595],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.245431,"threshold_uncertainty_score":0.7381353,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2414846825066956,"score_gpt":0.3999809639383797,"score_spread":0.1584962814316841,"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."}}