{"id":"W4415469702","doi":"10.54254/2753-8818/2025.dl27968","title":"Exploring the Trade-off Between Accuracy and Transparency in Credit Risk Prediction Models","year":2025,"lang":"","type":"article","venue":"Theoretical and Natural Science","topic":"Financial Distress and Bankruptcy Prediction","field":"Business, Management and Accounting","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"","keywords":"Interpretability; Logistic regression; Random forest; Decision tree; Boosting (machine learning); Credit risk; Predictive modelling; Metric (unit); Default; Ordinary least squares","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":["sts"],"consensus_categories":[],"category_scores_codex":[0.001544446,0.0002663835,0.00028679,0.000291394,0.00121163,0.000658266,0.0004571667,0.0000921732,0.00002174983],"category_scores_gemma":[0.0005692253,0.0001748528,0.00005870553,0.001801603,0.003199327,0.003937738,0.0002977874,0.0006826373,0.000002398418],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00004404458,"about_ca_system_score_gemma":0.00006350207,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0002803681,"about_ca_topic_score_gemma":0.00003067554,"domain_scores_codex":[0.9977538,0.00004888441,0.0005027899,0.0006629724,0.0004694606,0.0005620767],"domain_scores_gemma":[0.9992113,0.0003323412,0.0001283244,0.0002146594,0.0000712041,0.0000421621],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"observational","study_design_scores_codex":[0.0001038005,0.00003731545,0.02214509,0.0001210665,0.000008508244,0.000001908035,0.000296986,0.00004410756,0.0001027095,0.6437327,0.00002721275,0.3333786],"study_design_scores_gemma":[0.000466466,0.00002918502,0.6785547,0.0004051832,0.0001408843,7.815431e-7,0.0002553187,0.1234196,0.00007533841,0.1958725,0.0005966455,0.0001833352],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9865687,0.003237617,0.0008323391,0.004272272,0.001177174,0.0004220667,0.0000373704,0.00005117146,0.003401275],"genre_scores_gemma":[0.9942522,0.004963589,0.00001600254,0.0001718486,0.0005384007,0.00002746378,0.000006657559,0.000007070833,0.00001675445],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.6564096,"threshold_uncertainty_score":0.9995134,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02272191441987632,"score_gpt":0.2365992083686367,"score_spread":0.2138772939487603,"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."}}