{"id":"W4226003933","doi":"10.18653/v1/2022.acl-long.96","title":"Better Language Model with Hypernym Class Prediction","year":2022,"lang":"en","type":"article","venue":"Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)","topic":"Topic Modeling","field":"Computer Science","cited_by":9,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"Microsoft Research","keywords":"Perplexity; Computer science; Artificial intelligence; Transformer; Class (philosophy); Language model; WordNet; Security token; Natural language processing; Machine learning; Context (archaeology)","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":[0.002280987,0.0008892023,0.0008105711,0.001038102,0.0004108803,0.001415289,0.001327165,0.0009640639,0.002312326],"category_scores_gemma":[0.0065876,0.0003177853,0.001121485,0.0008632204,0.0003264367,0.00494475,0.0009765923,0.002731551,0.001786992],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0007298545,"about_ca_system_score_gemma":0.001024809,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00588125,"about_ca_topic_score_gemma":0.009560985,"domain_scores_codex":[0.9993122,0.0002964432,0.00003567539,0.0002422198,0.00006409108,0.00004947897],"domain_scores_gemma":[0.9977266,0.001422714,0.00008097592,0.0003884465,0.00030354,0.00007765708],"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":[0.0006453694,0.0004539587,0.01043965,0.0002136009,0.0002969131,0.0001792715,0.0005545094,0.5271757,0.01256514,0.015248,0.01433115,0.4178967],"study_design_scores_gemma":[0.000008577635,0.00001567119,0.0001406547,0.000004270629,0.000009671394,0.000008188505,0.00001277832,0.9947524,0.0007052262,0.003897091,0.0004406521,0.000004695609],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.2878604,0.00110962,0.6947632,0.002214736,0.0003343703,0.0001531618,0.001551862,0.007880093,0.0041326],"genre_scores_gemma":[0.8723606,0.0002344772,0.1187106,0.0004384486,0.0001701586,0.0001530694,0.002311609,0.0003759244,0.005245093],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.00588125,"threshold_uncertainty_score":0.01206315,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.006999110390969953,"score_gpt":0.2098802128432188,"score_spread":0.2028811024522488,"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."}}