{"id":"W6929034863","doi":"10.48448/6gek-4x60","title":"Director: Generator-Classifiers For Supervised Language Modeling","year":2022,"lang":"en","type":"other","venue":"Underline Science Inc.","topic":"","field":"","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University","funders":"","keywords":"Perplexity; Language model; Language understanding; Natural language; Language identification; Data modeling; Decoding methods; Cache language model","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","insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.001577928,0.0005632528,0.0005562011,0.00165234,0.000684403,0.0002580918,0.001910958,0.0002207149,0.009620379],"category_scores_gemma":[0.0003444161,0.0005507329,0.0001802337,0.001750307,0.0006914253,0.000231483,0.0004804697,0.0004490782,0.0006795684],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001027786,"about_ca_system_score_gemma":0.001469677,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0005342804,"about_ca_topic_score_gemma":0.001203167,"domain_scores_codex":[0.9953569,0.00009263235,0.0004260703,0.001459633,0.001530442,0.001134278],"domain_scores_gemma":[0.9979708,0.00007610732,0.0002603862,0.001188236,0.0001487272,0.0003557303],"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":[0.00004974628,0.000258673,0.0001072063,0.000163112,0.0001343671,0.00003470671,0.001133474,0.009647241,0.09508388,0.004576501,0.8817127,0.007098381],"study_design_scores_gemma":[0.0007449746,0.00007150156,7.037937e-7,0.00004013267,0.00006913109,0.000006345211,0.001267788,0.6569237,0.0005739763,0.00009036093,0.3394274,0.0007839712],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"other","genre_gemma":"other","genre_scores_codex":[0.007858024,0.006353865,0.02212971,0.0004953601,0.005778988,0.004924179,0.007011134,0.005534679,0.939914],"genre_scores_gemma":[0.06605124,0.0001101716,0.107945,0.001510342,0.004209515,0.001131948,0.002934266,0.006837007,0.8092705],"genre_candidate":"other","genre_consensus":"other","teacher_disagreement_score":0.6472764,"threshold_uncertainty_score":0.9996944,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04335961673155902,"score_gpt":0.3130141114714701,"score_spread":0.2696544947399111,"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."}}