{"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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.003650613,0.002971209,0.001055961,0.001604398,0.0006068228,0.003542541,0.00446282,0.001210873,0.01380445],"category_scores_gemma":[0.01785934,0.001233545,0.002743761,0.001618255,0.00118309,0.007143198,0.004461485,0.00338337,0.009691223],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001191685,"about_ca_system_score_gemma":0.003398542,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.006926826,"about_ca_topic_score_gemma":0.01212657,"domain_scores_codex":[0.99632,0.0007988986,0.0005303076,0.00105762,0.001071194,0.0002219581],"domain_scores_gemma":[0.9923707,0.003035181,0.0003312786,0.002963106,0.0009779618,0.0003218011],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00227742,0.001428899,0.01328332,0.002088995,0.000846323,0.0009614643,0.0009966422,0.09321193,0.03979781,0.029908,0.3199873,0.495212],"study_design_scores_gemma":[0.0003558803,0.0002434401,0.001297036,0.00009580379,0.00008446234,0.0003156716,0.0002962023,0.886759,0.02797657,0.02303201,0.05941276,0.0001311821],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.02701768,0.0005717836,0.5851592,0.0008995058,0.0004486222,0.0008936138,0.01003342,0.3701124,0.004863766],"genre_scores_gemma":[0.2402729,0.0005787119,0.6507499,0.001953181,0.0001957748,0.001037244,0.06292537,0.03280559,0.009481288],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.01380445,"threshold_uncertainty_score":0.04618055,"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."}}