{"id":"W7105692598","doi":"10.1109/ijcnn64981.2025.11227916","title":"Model Cascading for Code: A Cascaded Black-Box Multi-Model Framework for Cost-Efficient Code Completion with Self-Testing","year":2025,"lang":"","type":"article","venue":"","topic":"Software Testing and Debugging Techniques","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"York University","keywords":"Code (set theory); Heuristic; Server; Sensitivity (control systems); Test case; Source code; Computational complexity theory; Component (thermodynamics)","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow","sts"],"consensus_categories":[],"category_scores_codex":[0.002186927,0.00104069,0.001165985,0.0006174346,0.001789822,0.0009732448,0.001837567,0.0006733236,0.000001991433],"category_scores_gemma":[0.003976955,0.0009832361,0.00037867,0.00177435,0.0003365848,0.000361468,0.0007153862,0.0007999334,0.000005272311],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0007736317,"about_ca_system_score_gemma":0.001395342,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001056706,"about_ca_topic_score_gemma":0.00007354106,"domain_scores_codex":[0.9935891,0.0001306271,0.001357997,0.002378091,0.0007130112,0.001831176],"domain_scores_gemma":[0.9885392,0.006487882,0.0007247223,0.001704179,0.002154634,0.0003893564],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0002372078,0.0009107943,0.0007067014,0.0009108685,0.0001719966,0.000007317265,0.002756653,0.9003216,0.0001892665,0.07562223,0.008277211,0.009888143],"study_design_scores_gemma":[0.002530251,0.0005270766,0.00004638401,0.002441332,0.0002960603,0.00004321798,0.00005093022,0.9314583,0.001406486,0.05991163,0.0001528546,0.00113546],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.004322262,0.0001671155,0.9705383,0.001428709,0.0004162943,0.007303059,0.0002301694,0.01533858,0.000255481],"genre_scores_gemma":[0.3748362,0.000006675844,0.6220284,0.001277896,0.00006783652,0.001242864,0.00001592024,0.00008677164,0.0004374377],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.3705139,"threshold_uncertainty_score":0.9995097,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1273860547350612,"score_gpt":0.3602348113149539,"score_spread":0.2328487565798927,"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."}}