{"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":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.001136319,0.0003791056,0.0004156362,0.0004887039,0.0004756373,0.0003598015,0.004002747,0.0002709729,0.0001967365],"category_scores_gemma":[0.0008832876,0.0003790277,0.00006747104,0.001796086,0.0002822251,0.0009804068,0.001344543,0.000318168,0.00001726108],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002605117,"about_ca_system_score_gemma":0.001523234,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0006062334,"about_ca_topic_score_gemma":0.0005620756,"domain_scores_codex":[0.9963666,0.00006243872,0.0004399295,0.001753084,0.000549946,0.0008279964],"domain_scores_gemma":[0.9968231,0.0001426879,0.0004008384,0.001543896,0.0009164016,0.0001730695],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00002699135,0.0002750759,0.0000515656,0.0007552502,0.0001330017,0.00001651768,0.000353419,0.07721601,0.002179168,0.01921639,0.7897059,0.1100707],"study_design_scores_gemma":[0.0002551611,0.00003707664,3.191196e-7,0.0002442464,0.00002166318,0.00001186833,0.00006862144,0.6940556,0.0006167983,0.01416534,0.2901748,0.0003485311],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.000005927084,0.000161787,0.7311311,0.0015351,0.001238949,0.0006292037,0.0002145672,0.000619521,0.2644638],"genre_scores_gemma":[0.003519807,0.0002283472,0.691044,0.000665079,0.001241142,0.0001095648,0.001343275,0.00006314963,0.3017856],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.6168395,"threshold_uncertainty_score":0.9998662,"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."}}