{"id":"W4388878572","doi":"10.1007/978-3-031-48312-7_36","title":"Multi-task Learning over Mixup Variants for the Speaker Verification Task","year":2023,"lang":"en","type":"book-chapter","venue":"Lecture notes in computer science","topic":"Speech Recognition and Synthesis","field":"Computer Science","cited_by":3,"is_retracted":false,"has_abstract":false,"ca_institutions":"Computer Research Institute of Montréal","funders":"","keywords":"Computer science; Speech recognition; Robustness (evolution); Artificial intelligence; Regularization (linguistics); Machine learning","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":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.001453455,0.0004247175,0.0003723795,0.0005228237,0.0005878849,0.0007647174,0.002733956,0.0002845776,0.00004056479],"category_scores_gemma":[0.0005145677,0.0003246749,0.0002021888,0.0006390907,0.0003914088,0.0004384398,0.0006222497,0.0006776454,0.0002544747],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001664189,"about_ca_system_score_gemma":0.0002867982,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00002921108,"about_ca_topic_score_gemma":0.00008117255,"domain_scores_codex":[0.9966921,0.00005574668,0.000459323,0.001396173,0.0007996957,0.0005969361],"domain_scores_gemma":[0.9961811,0.001980371,0.0003048349,0.001150072,0.0002628797,0.0001207025],"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":[0.000007134226,0.00002492706,0.00003362291,0.0000242862,0.00002708372,0.00001971252,0.0006612765,0.005067525,0.0008197479,0.004463671,0.000125724,0.9887253],"study_design_scores_gemma":[0.0003557275,0.00007299677,0.001379404,0.0001914245,0.00001903562,0.00003019381,6.837724e-7,0.9662025,0.0006330251,0.01857404,0.01201367,0.0005273112],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.00002830095,0.0001367873,0.9943433,0.001122623,0.002691528,0.0006606111,0.00001316161,0.0002838229,0.0007198062],"genre_scores_gemma":[0.04820846,0.0002481634,0.9404415,0.003093911,0.001018887,0.0001257283,0.00002875847,0.0001286365,0.006705903],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.988198,"threshold_uncertainty_score":0.9999205,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04122870883359776,"score_gpt":0.2696667201367999,"score_spread":0.2284380113032021,"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."}}