{"id":"W4404088722","doi":"10.48550/arxiv.2410.15516","title":"Generating Tabular Data Using Heterogeneous Sequential Feature Forest Flow Matching","year":2024,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Time Series Analysis and Forecasting","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Institut de Valorisation des Données","keywords":"Matching (statistics); Feature (linguistics); Computer science; Feature matching; Flow (mathematics); Data mining; Artificial intelligence; Mathematics; Feature extraction; Statistics","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001361956,0.0006440603,0.0005839335,0.001036019,0.0005117533,0.0008544372,0.001073208,0.0008616368,0.005332112],"category_scores_gemma":[0.007213663,0.0003140305,0.001127948,0.001291079,0.0004649983,0.00132903,0.0009965008,0.0009066291,0.001065351],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000809823,"about_ca_system_score_gemma":0.001365447,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004377932,"about_ca_topic_score_gemma":0.005585749,"domain_scores_codex":[0.9994579,0.000147525,0.00003642846,0.0001556814,0.0001352794,0.00006716636],"domain_scores_gemma":[0.9975615,0.00141229,0.00014442,0.0004319436,0.0003696294,0.00008037491],"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":[0.0002567816,0.0001834142,0.00713751,0.0001623287,0.00005749859,0.0002018605,0.0001669461,0.6300837,0.00531594,0.02386485,0.01359578,0.3189733],"study_design_scores_gemma":[0.00002033058,0.00002029578,0.0002558834,0.000007210335,0.000003752163,0.00002405119,0.00001496944,0.9835274,0.001631167,0.01301652,0.001472385,0.00000618751],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.04700943,0.0001971706,0.9439045,0.0002806579,0.0001019812,0.0002218606,0.001988589,0.00422716,0.002068705],"genre_scores_gemma":[0.3808883,0.0001132954,0.6087996,0.0002375701,0.00005472587,0.0004766956,0.00648379,0.0006509856,0.002294966],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.005332112,"threshold_uncertainty_score":0.0178377,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1075741424326963,"score_gpt":0.2124400605726323,"score_spread":0.1048659181399359,"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."}}