{"id":"W7082251285","doi":"10.48448/n881-4936","title":"Advancing African-Accented English Speech Recognition: Epistemic Uncertainty-Driven Data Selection for Generalizable ASR Models","year":2025,"lang":"en","type":"other","venue":"Underline Science Inc.","topic":"Geochemistry and Geologic Mapping","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University; Mila - Quebec Artificial Intelligence Institute","funders":"","keywords":"CLARITY; Context (archaeology); Process (computing); Generalization; Adaptation (eye); Metric (unit); Selection (genetic algorithm); Annotation","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.004950606,0.0020395,0.001326095,0.001062117,0.0006629874,0.001819199,0.002273722,0.001605322,0.002815695],"category_scores_gemma":[0.01195534,0.0006058901,0.001306641,0.000762413,0.0009451304,0.002503563,0.003244962,0.003130744,0.003193728],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0008023737,"about_ca_system_score_gemma":0.001095298,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004937346,"about_ca_topic_score_gemma":0.006685517,"domain_scores_codex":[0.9973762,0.001156974,0.000147941,0.0007563773,0.0003960867,0.0001665102],"domain_scores_gemma":[0.9955013,0.002715362,0.0001939392,0.0007584009,0.0007078215,0.0001232322],"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":[0.0006907738,0.0003313544,0.004072948,0.0002282273,0.000228452,0.0003291937,0.0007457934,0.2201001,0.03119888,0.005768949,0.01036628,0.7259391],"study_design_scores_gemma":[0.00002579739,0.00006865705,0.0008375642,0.00002489897,0.00003385548,0.00008120115,0.0001114612,0.9769728,0.0113103,0.00741191,0.003090031,0.00003144653],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.03417185,0.0006221827,0.9542829,0.0006208497,0.0001272483,0.000145621,0.0007065829,0.007155479,0.002167283],"genre_scores_gemma":[0.5388023,0.0005290643,0.4444357,0.001044735,0.0002598132,0.0005766628,0.005801656,0.001317073,0.007232959],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.004950606,"threshold_uncertainty_score":0.02618164,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04416266560626426,"score_gpt":0.2855035908848412,"score_spread":0.2413409252785769,"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."}}