{"id":"W6910447145","doi":"10.48448/987h-0v73","title":"NormTab: Improving Symbolic Reasoning in LLMs Through Tabular Data Normalization","year":2024,"lang":"en","type":"other","venue":"Underline Science Inc.","topic":"","field":"","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta","funders":"","keywords":"Normalization (sociology); Preprocessor; Parsing; The Symbolic; Table (database); Parse tree; Symbolic data analysis","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":[0.004357232,0.001969817,0.001147841,0.003280578,0.001046672,0.004582543,0.004126112,0.001065969,0.008607341],"category_scores_gemma":[0.03404965,0.0008266825,0.002444511,0.003730572,0.001629657,0.008040631,0.004898902,0.002863306,0.005140292],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00217103,"about_ca_system_score_gemma":0.006377745,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0124957,"about_ca_topic_score_gemma":0.02124254,"domain_scores_codex":[0.992724,0.002105377,0.0009774766,0.001345901,0.002558465,0.0002888513],"domain_scores_gemma":[0.9847125,0.006302699,0.001071807,0.00544987,0.002196852,0.000266253],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.0005634244,0.0005086209,0.009205666,0.001440138,0.0002724045,0.0004270418,0.001048356,0.04078001,0.0232663,0.045782,0.08836854,0.7883376],"study_design_scores_gemma":[0.0001418483,0.0001281571,0.001709609,0.0002691775,0.0001413473,0.0003256091,0.0005372192,0.7406937,0.07128569,0.1064756,0.07813711,0.0001548376],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01161231,0.0007042683,0.8277274,0.001014665,0.0002255732,0.0002410695,0.005131893,0.1503553,0.002987549],"genre_scores_gemma":[0.0854494,0.0003646513,0.8881071,0.0007671744,0.0000895496,0.0003520469,0.01562572,0.006294742,0.002949566],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.0124957,"threshold_uncertainty_score":0.02879441,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04501706533176654,"score_gpt":0.3431838697605389,"score_spread":0.2981668044287724,"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."}}