{"id":"W4388963821","doi":"10.48550/arxiv.2311.13517","title":"Learning-Based Relaxation of Completeness Requirements for Data Entry Forms","year":2023,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Data Quality and Management","field":"Decision Sciences","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Ottawa","funders":"BNP Paribas Cardif","keywords":"Completeness (order theory); Computer science; Field (mathematics); Lacquer; Data mining; Artificial intelligence; Machine learning; Mathematics","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":[],"consensus_categories":[],"category_scores_codex":[0.004594393,0.0002178336,0.0004598618,0.0004915736,0.0001925014,0.000137177,0.003827339,0.0001866768,0.00009633],"category_scores_gemma":[0.002461443,0.0002128834,0.0001911628,0.0007222203,0.0001506675,0.0005520416,0.004421141,0.0002633778,0.0001745086],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001034119,"about_ca_system_score_gemma":0.0001683848,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0002781754,"about_ca_topic_score_gemma":0.0002747366,"domain_scores_codex":[0.9969301,0.0002934066,0.0006456584,0.001381368,0.0004674243,0.000282068],"domain_scores_gemma":[0.9943928,0.001145836,0.001189042,0.002776468,0.0004029236,0.00009288932],"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":[0.0006132597,0.0003632637,0.01812363,0.0006896928,0.0003395371,0.00004790324,0.0002350895,0.833593,0.00002951198,0.09450283,0.04561412,0.005848213],"study_design_scores_gemma":[0.001353532,0.0001467277,0.004274046,0.0001827278,0.0002004268,1.258837e-7,0.00117669,0.7703267,0.00007492047,0.1727818,0.04908618,0.0003961496],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.09276608,0.00001125208,0.9020838,0.0004663442,0.0009197789,0.0009652532,0.001445762,0.000145118,0.001196652],"genre_scores_gemma":[0.9867567,0.00005026712,0.0005315418,0.00009778328,0.00003887346,0.000002741096,0.003398244,0.00002133013,0.009102473],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9015522,"threshold_uncertainty_score":0.8681132,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.649970621585224,"score_gpt":0.3571455270505782,"score_spread":0.2928250945346458,"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."}}