{"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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.002532372,0.001229183,0.0007235846,0.001168258,0.0006090623,0.001071229,0.002517355,0.00143173,0.009523857],"category_scores_gemma":[0.007444721,0.000768323,0.0007438503,0.0007226607,0.0005587781,0.002689122,0.001764165,0.002727494,0.008601804],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0009025034,"about_ca_system_score_gemma":0.001731046,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003229839,"about_ca_topic_score_gemma":0.008098498,"domain_scores_codex":[0.9984968,0.0005559559,0.00007478336,0.0004391848,0.0003443875,0.00008900644],"domain_scores_gemma":[0.9976342,0.001187784,0.00009860967,0.0005534628,0.0004223863,0.0001036124],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"not_applicable","study_design_scores_codex":[0.00032157,0.0002247172,0.001693876,0.0002330517,0.0001247873,0.0001719501,0.0001702198,0.06629092,0.01101374,0.02577375,0.09619378,0.7977876],"study_design_scores_gemma":[0.00006208776,0.00006082043,0.0002365862,0.00002682957,0.0000218008,0.0001024283,0.0000244382,0.9367531,0.01363818,0.02724642,0.02180198,0.00002528392],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"software","genre_scores_codex":[0.003674721,0.000227812,0.9632849,0.0002408484,0.000127945,0.0001174893,0.0008532488,0.02917017,0.002302999],"genre_scores_gemma":[0.1490615,0.0003789068,0.8181871,0.0006238657,0.0002313875,0.0007486973,0.008849842,0.003560432,0.01835821],"genre_candidate":"software","genre_consensus":null,"teacher_disagreement_score":0.009523857,"threshold_uncertainty_score":0.03186053,"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."}}