{"id":"W4404783544","doi":"10.18653/v1/2024.emnlp-main.302","title":"Improving Spoken Language Modeling with Phoneme Classification: A Simple Fine-tuning Approach","year":2024,"lang":"en","type":"article","venue":"","topic":"Speech Recognition and Synthesis","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Agence de l'innovation de Défense; Grand Équipement National De Calcul Intensif; Agence Nationale de la Recherche; École des Hautes Etudes en Sciences Sociales; Canadian Institute for Advanced Research","keywords":"Computer science; Simple (philosophy); Spoken language; Speech recognition; Natural language processing; Language model; Artificial intelligence; Fine-tuning","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0002359603,0.0001173589,0.0001068214,0.0001387991,0.00009889554,0.0004670283,0.000349397,0.00004197569,0.0001226369],"category_scores_gemma":[0.00002219221,0.0000863665,0.00004743193,0.0004488883,0.00001465613,0.000470158,0.00008168943,0.0001238222,0.0001171437],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00003289855,"about_ca_system_score_gemma":0.00007556185,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001018245,"about_ca_topic_score_gemma":0.00002223048,"domain_scores_codex":[0.9989786,0.00002639981,0.0001538155,0.0004133402,0.0002202208,0.0002076545],"domain_scores_gemma":[0.9994463,0.00006354808,0.00002329928,0.0003493387,0.00004424313,0.00007324018],"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":[0.000007144416,0.00005984105,0.00003322972,0.0001109043,0.00004621603,0.00007645103,0.00251817,0.0006205826,0.01090311,0.05033375,0.0006373794,0.9346532],"study_design_scores_gemma":[0.00008543525,0.00001446399,0.00001200867,0.00002189479,0.000007219361,0.00004682452,0.000773628,0.9967368,0.001580998,0.0002264015,0.0003432347,0.0001510878],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01328855,0.0001372938,0.9468845,0.0003542648,0.00006580498,0.0001028855,0.000001525917,0.0007115169,0.03845372],"genre_scores_gemma":[0.6882848,0.000002189247,0.3107153,0.0001326725,0.00007112572,0.00002387136,0.000007157409,0.00001208991,0.0007508492],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9961162,"threshold_uncertainty_score":0.4503565,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03837804051959954,"score_gpt":0.2513076571257065,"score_spread":0.2129296166061069,"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."}}