{"id":"W4403863506","doi":"10.1109/taslp.2024.3487410","title":"CL-MASR: A Continual Learning Benchmark for Multilingual ASR","year":2024,"lang":"en","type":"article","venue":"IEEE/ACM Transactions on Audio Speech and Language Processing","topic":"Speech Recognition and Synthesis","field":"Computer Science","cited_by":6,"is_retracted":false,"has_abstract":true,"ca_institutions":"Université Laval; Concordia University","funders":"Natural Sciences and Engineering Research Council of Canada; Alliance de recherche numérique du Canada","keywords":"Benchmark (surveying); Computer science; Artificial intelligence; Geography","routes":{"ca_aff":true,"ca_fund":true,"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":[0.0004171568,0.000250486,0.000249427,0.000318071,0.0004735704,0.0007442975,0.0003451634,0.0001346198,0.0001296772],"category_scores_gemma":[0.0001335636,0.0002289279,0.0001576653,0.0004391694,0.00006403284,0.000550625,0.000009643331,0.0004180378,0.00004283739],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00004682925,"about_ca_system_score_gemma":0.0001288298,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00002039756,"about_ca_topic_score_gemma":0.00004391715,"domain_scores_codex":[0.9983436,0.00006127085,0.0002901385,0.0006358745,0.0002643457,0.0004047472],"domain_scores_gemma":[0.9989816,0.0004327771,0.00006274199,0.0002669615,0.00009939404,0.0001565152],"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":[0.00002488652,0.00005771801,0.000004432453,0.0001676535,0.00003657906,0.00007927037,0.003599514,0.0000320732,0.007196973,0.00002331774,0.00007665654,0.9887009],"study_design_scores_gemma":[0.002158332,0.000630117,0.00008077669,0.001568181,0.0002285942,0.0007812689,0.006571142,0.3227603,0.6301467,0.001479462,0.03207556,0.001519532],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.05716243,0.002518234,0.9365783,0.000897426,0.0006019091,0.0003222629,0.00002564107,0.0008299385,0.001063792],"genre_scores_gemma":[0.8171898,0.00008619169,0.1794992,0.0003473005,0.0002073129,0.00007723238,0.000007157308,0.00003735873,0.002548452],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9871814,"threshold_uncertainty_score":0.9335409,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01732500977189165,"score_gpt":0.2885879270057755,"score_spread":0.2712629172338838,"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."}}