{"id":"W4401023759","doi":"10.24963/ijcai.2024/572","title":"Optimal Transport on Categorical Data for Conterfactuals Using Compositional Data and Dirichlet Transport","year":2024,"lang":"en","type":"article","venue":"","topic":"Speech and Audio Processing","field":"Computer Science","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"Université de Montréal; Université du Québec à Montréal","funders":"Office of Advanced Cyberinfrastructure; National Centre for Supercomputing Applications; National Science Foundation","keywords":"Computer science; Speech recognition; Microphone; Discriminative model; Speaker diarisation; Speaker recognition; Leverage (statistics); Context (archaeology); Artificial intelligence; Task (project management); Unsupervised learning; Reduction (mathematics); Speech enhancement; Noise reduction; Pattern recognition (psychology); 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":[0.0003341433,0.0001539883,0.0001849111,0.00007482708,0.0001392275,0.0002081499,0.001383532,0.0000522109,0.00002372459],"category_scores_gemma":[0.000007684324,0.0001246047,0.00002686013,0.0001618538,0.00005324791,0.001395258,0.0001721492,0.0001158081,0.000005631658],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00001790579,"about_ca_system_score_gemma":0.0001511114,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001806329,"about_ca_topic_score_gemma":0.000006457714,"domain_scores_codex":[0.9983941,0.00001089049,0.0002390575,0.0008577195,0.0002449448,0.0002532871],"domain_scores_gemma":[0.998828,0.0001508535,0.00002481413,0.0008592904,0.00002923165,0.0001077888],"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":[0.001207431,0.002548326,0.008712503,0.0035086,0.001881042,0.00314972,0.006314118,0.007179133,0.1904235,0.1401258,0.08748382,0.547466],"study_design_scores_gemma":[0.0006576723,0.0001675544,0.001094267,0.0001329159,0.00008146693,0.0002427294,0.00003962205,0.9474549,0.01730529,0.00106625,0.03129199,0.0004653206],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.06163402,0.000500154,0.9343842,0.001922301,0.0002224494,0.0001604796,0.0007024378,0.0001800577,0.0002939391],"genre_scores_gemma":[0.7810989,0.00001385447,0.2172788,0.0005242247,0.0001311332,0.000003092496,0.0008439927,0.00001241378,0.00009355649],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9402758,"threshold_uncertainty_score":0.5081233,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1231895037104952,"score_gpt":0.3489885142031081,"score_spread":0.2257990104926129,"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."}}