{"id":"W4404782579","doi":"10.18653/v1/2024.findings-emnlp.481","title":"DTS-SQL: Decomposed Text-to-SQL with Small Large Language Models","year":2024,"lang":"en","type":"article","venue":"","topic":"Scientific Computing and Data Management","field":"Decision Sciences","cited_by":47,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Computer science; SQL; SQL/PSM; Programming language; PL/SQL; Stored procedure; Language Integrated Query; Data definition language; Query by Example; SQL injection; Natural language processing; Database; Information retrieval; Search engine","routes":{"ca_aff":true,"ca_fund":true,"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":["scholarly_communication","insufficient_payload"],"consensus_categories":["insufficient_payload"],"category_scores_codex":[0.004431256,0.0001771251,0.000216992,0.0005392255,0.0001744776,0.001989113,0.001294168,0.00003640248,0.001402592],"category_scores_gemma":[0.0002886482,0.0001045688,0.00008889201,0.001880697,0.00003791798,0.000345697,0.0007494559,0.0001224376,0.005163471],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00003416269,"about_ca_system_score_gemma":0.00006570303,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001072447,"about_ca_topic_score_gemma":0.0006595604,"domain_scores_codex":[0.9965498,0.00008943515,0.0004263266,0.001170315,0.001303345,0.0004608067],"domain_scores_gemma":[0.9975573,0.0006319828,0.000044188,0.001418917,0.000120084,0.0002275608],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"not_applicable","study_design_scores_codex":[0.00005562233,0.0001919886,0.0003631539,0.00002712998,0.00007329972,0.0003537432,0.005131995,0.01084098,0.0005153142,0.07594195,0.5363125,0.3701923],"study_design_scores_gemma":[0.0003523588,0.00009469326,0.000727277,0.00007100224,0.00002072214,0.00001534853,0.003509117,0.4562967,0.0002208285,0.008571844,0.5297608,0.000359285],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1121212,0.0002153325,0.7672429,0.002384797,0.001187673,0.0003234043,0.0001170823,0.0005695029,0.115838],"genre_scores_gemma":[0.8883491,0.000001421529,0.02655822,0.001417165,0.0001151461,0.00001037517,0.00002608446,0.00002081093,0.0835017],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.7762278,"threshold_uncertainty_score":0.9995103,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1252461638605778,"score_gpt":0.3780946688147803,"score_spread":0.2528485049542025,"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."}}