{"id":"W7125586192","doi":"10.1109/cascon66301.2025.00111","title":"ScenarioBench: Trace-Grounded Compliance Evaluation for Text-to-SQL and RAG","year":2025,"lang":"","type":"article","venue":"","topic":"Scientific Computing and Data Management","field":"Decision Sciences","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Ontario Tech University","funders":"","keywords":"Correctness; Benchmark (surveying); Artifact (error); Code (set theory); Matching (statistics); Aggregate (composite); Witness; Core (optical fiber); Compliance (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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.009730566,0.001997112,0.0007191059,0.002234233,0.0006549216,0.002431934,0.003532783,0.002192728,0.008167396],"category_scores_gemma":[0.05373091,0.0007268729,0.001046875,0.001652101,0.001511608,0.003771196,0.002842489,0.002088624,0.002037514],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002136715,"about_ca_system_score_gemma":0.002725407,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.007211014,"about_ca_topic_score_gemma":0.007778744,"domain_scores_codex":[0.9852855,0.007693483,0.001221596,0.001763934,0.003289772,0.0007457341],"domain_scores_gemma":[0.9661374,0.01977928,0.00147292,0.008038255,0.003321897,0.001250146],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.006527802,0.004234145,0.03253324,0.002617427,0.0008244528,0.001327357,0.001755569,0.4618834,0.03299894,0.0373102,0.1159264,0.302061],"study_design_scores_gemma":[0.0005344915,0.001145738,0.004089168,0.00009640176,0.00005718275,0.000239626,0.0005053252,0.9394134,0.02663087,0.01479863,0.01240446,0.00008476068],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"methods","genre_scores_codex":[0.5154594,0.0009326701,0.3103442,0.001206672,0.000459634,0.001860684,0.02004668,0.1278332,0.02185687],"genre_scores_gemma":[0.7259823,0.0001651566,0.2404281,0.0003604656,0.00003910722,0.0006160382,0.02441993,0.005264322,0.002724566],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.009730566,"threshold_uncertainty_score":0.05146074,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.316861616734726,"score_gpt":0.4856278158292047,"score_spread":0.1687661990944787,"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."}}