{"id":"W4408401921","doi":"10.1109/icftic64248.2024.10913131","title":"A Modular Agent-Based Approach to Complex Data Question-Answering with SQL Generation and Parameter-Efficient Fine-Tuning","year":2024,"lang":"en","type":"article","venue":"","topic":"Data Management and Algorithms","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Simon Fraser University","funders":"","keywords":"Modular design; Computer science; SQL; Question answering; Stored procedure; Programming language; Artificial intelligence; Database; Information retrieval; Query by Example; Search engine","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":"codex-gemma-dda1882f352a","candidate_categories":["scholarly_communication"],"consensus_categories":[],"category_scores_codex":[0.000370482,0.0001286647,0.00009958931,0.0001507047,0.0001081929,0.00116089,0.000699009,0.0000166411,0.000006740724],"category_scores_gemma":[0.00001386209,0.00009999988,0.00001190761,0.0004227741,0.00002051882,0.0005796844,0.0006227315,0.0000606481,0.00001107095],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00001943779,"about_ca_system_score_gemma":0.00002281115,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00006430358,"about_ca_topic_score_gemma":0.00001514266,"domain_scores_codex":[0.9986976,0.00003410539,0.0001346876,0.0006936686,0.0002506016,0.0001893612],"domain_scores_gemma":[0.9990013,0.00002937488,0.00001823536,0.0008470372,0.00002196724,0.00008206934],"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":[0.00001001237,0.0003078969,0.0001638678,0.0002816267,0.0001280921,0.00009588432,0.0006399387,0.4768947,0.006488941,0.1679584,0.01294565,0.334085],"study_design_scores_gemma":[0.0001169601,0.00004831257,0.0001872751,0.00003134682,0.00001189049,0.000004189931,0.00001445532,0.9927652,0.0001429451,0.00002211357,0.006495678,0.0001595654],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.009650783,0.00006649794,0.9881554,0.0006422903,0.0001273896,0.0002535973,0.0000218859,0.0002748282,0.0008072891],"genre_scores_gemma":[0.4088457,0.000001388704,0.5902909,0.0002394373,0.00005876388,0.00001836943,0.0003213503,0.00000882655,0.0002152653],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.5158706,"threshold_uncertainty_score":0.999876,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.09667619869056282,"score_gpt":0.288333229888341,"score_spread":0.1916570311977782,"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."}}