{"id":"W4404752327","doi":"10.18653/v1/2022.aacl-main.39","title":"Director: Generator-Classifiers For Supervised Language Modeling","year":2022,"lang":"en","type":"article","venue":"","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":11,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University","funders":"","keywords":"Computer science; Generator (circuit theory); Artificial intelligence; Language model; Natural language processing; Physics; Power (physics)","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":[],"consensus_categories":[],"category_scores_codex":[0.0003029851,0.0001070085,0.0001140596,0.0001045581,0.0003266327,0.000125594,0.0009962221,0.00002854016,0.00006761235],"category_scores_gemma":[0.00003440694,0.00009665632,0.00007166936,0.0002832366,0.000009196178,0.0002707251,0.0004536449,0.0001435854,0.000002269388],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00009487919,"about_ca_system_score_gemma":0.00006847124,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00004652864,"about_ca_topic_score_gemma":0.000007061581,"domain_scores_codex":[0.998985,0.0000451594,0.0001408583,0.0003390412,0.0002414992,0.0002484408],"domain_scores_gemma":[0.9994315,0.00004227442,0.00003180499,0.0003898904,0.00004838079,0.00005616431],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00004624488,0.0001816452,0.000096215,0.00009473777,0.00005898506,0.00006606711,0.006687766,0.003802839,0.3404804,0.4537767,0.02749467,0.1672136],"study_design_scores_gemma":[0.0001854265,0.00004814522,2.924642e-7,0.000002437243,0.000003184016,0.000008953039,0.000185862,0.9498958,0.04152445,0.006163564,0.001792454,0.0001894542],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.01390803,0.001884276,0.9809771,0.0008131765,0.0002465878,0.0002186878,0.000008359227,0.001294419,0.000649361],"genre_scores_gemma":[0.4833482,0.000001475672,0.5151137,0.0007865029,0.00004541294,0.0001354103,0.00000730588,0.00001113099,0.000550852],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.946093,"threshold_uncertainty_score":0.3941531,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02489321284759761,"score_gpt":0.2742018675418575,"score_spread":0.2493086546942599,"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."}}