{"id":"W4407221739","doi":"10.48550/arxiv.2502.03128","title":"Metis: A Foundation Speech Generation Model with Masked Generative Pre-training","year":2025,"lang":"en","type":"preprint","venue":"ArXiv.org","topic":"Speech and dialogue systems","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Metis; Generative grammar; Foundation (evidence); Training (meteorology); Speech recognition; Computer science; Psychology; Artificial intelligence; History; Geography; World Wide Web","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":false,"about_ca":true,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0005232648,0.0004254461,0.000480188,0.0002641915,0.0002305176,0.000450394,0.0009751508,0.0003153451,0.000009720665],"category_scores_gemma":[0.00008497304,0.0003798852,0.0001230013,0.000382204,0.0000524512,0.0005091637,0.0007082318,0.0004637794,0.00006301526],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002211205,"about_ca_system_score_gemma":0.001144689,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0002414048,"about_ca_topic_score_gemma":0.000385904,"domain_scores_codex":[0.9972182,0.0002078425,0.0005000695,0.001176571,0.0004880073,0.0004092957],"domain_scores_gemma":[0.9979025,0.00005630184,0.0003684292,0.001202723,0.0003446192,0.0001254527],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0002398885,0.0005151116,0.05190467,0.0008519472,0.001563225,0.0002146292,0.04507167,0.6813217,0.03654639,0.05451818,0.005896079,0.1213565],"study_design_scores_gemma":[0.0005511073,0.00009699557,0.002622414,0.0002328949,0.00005792652,0.00001483313,0.0000676144,0.9790391,0.01329609,0.00303009,0.0003984618,0.0005924171],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2857731,0.0001065432,0.7087205,0.000462137,0.001143302,0.0007035381,0.00001531984,0.0002742919,0.0028012],"genre_scores_gemma":[0.7874252,0.00002240387,0.2075919,0.000438298,0.0007630754,0.0003170677,0.0002814055,0.00002293094,0.00313776],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.5016521,"threshold_uncertainty_score":0.9998653,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.09950370094731856,"score_gpt":0.298433028713288,"score_spread":0.1989293277659695,"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."}}