{"id":"W7134954916","doi":"10.1109/icdmw69685.2025.00343","title":"Enhancing LLM Fine-Tuning for Text-to-SQLs by SQL Quality Measurement","year":2025,"lang":"","type":"article","venue":"","topic":"SAS software applications and methods","field":"Engineering","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Quality (philosophy); SQL; Measure (data warehouse); Data collection; Work (physics); Data quality","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"],"consensus_categories":[],"category_scores_codex":[0.003912879,0.0004339024,0.0005935086,0.0002219074,0.0003393845,0.0001413741,0.0004022478,0.0002356648,0.0003220766],"category_scores_gemma":[0.001476725,0.0004775135,0.0002423013,0.0006876532,0.00002908239,0.0001155864,0.0001401769,0.0002538289,0.00006860978],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0006185545,"about_ca_system_score_gemma":0.0002161162,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0003204479,"about_ca_topic_score_gemma":0.0003679738,"domain_scores_codex":[0.9969004,0.0001231709,0.001113727,0.0007052456,0.0004032832,0.0007541281],"domain_scores_gemma":[0.9973226,0.0009375992,0.00009841687,0.0008917326,0.0004964289,0.0002531907],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"not_applicable","study_design_scores_codex":[0.00004610124,0.0002175992,0.0001835985,0.0009988605,0.0003466542,2.051856e-7,0.0006826802,0.001966109,0.5854809,0.0167627,0.03758851,0.355726],"study_design_scores_gemma":[0.00110073,0.0001004925,0.0005685325,0.0004416183,0.0001889298,5.393692e-7,0.0005421598,0.006329583,0.4788861,0.002034015,0.509013,0.000794331],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.002553443,0.00237735,0.9773921,0.001136149,0.0008254602,0.001671548,0.00005463602,0.0004331698,0.01355611],"genre_scores_gemma":[0.417563,0.00007826568,0.5691314,0.001043484,0.0002110538,0.001429823,0.00002249306,0.00009139471,0.01042914],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.4714245,"threshold_uncertainty_score":0.9997677,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05747556567261453,"score_gpt":0.3639704449943609,"score_spread":0.3064948793217464,"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."}}