{"id":"W7104018988","doi":"10.18419/opus-17489","title":"Can BERT generate Quebec English? : exploring and improving the acceptability of regional variation in pretrained language models","year":2024,"lang":"en","type":"other","venue":"OPUS Publication Server of the University of Stuttgart (University of Stuttgart)","topic":"","field":"","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Variety (cybernetics); Language model; Variation (astronomy); Sentence; Natural language; Natural (archaeology); Sentence processing; On Language","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.00156388,0.001579394,0.0004719383,0.0005907504,0.0007505192,0.001478227,0.001659262,0.000747863,0.004466098],"category_scores_gemma":[0.00667608,0.0004628029,0.0006621621,0.0004568012,0.000460661,0.002005085,0.001001093,0.001887537,0.001880764],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002694602,"about_ca_system_score_gemma":0.00316109,"about_ca_topic_candidate":true,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.3352661,"about_ca_topic_score_gemma":0.4517112,"domain_scores_codex":[0.9993938,0.0002257747,0.00002146104,0.0002380234,0.00005798019,0.00006289658],"domain_scores_gemma":[0.9972223,0.001767073,0.00008033947,0.0003278468,0.0004821053,0.0001203791],"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":[0.001509581,0.000640317,0.03900149,0.000436919,0.0005231885,0.0008937909,0.001943012,0.4834301,0.02569265,0.006633171,0.04028279,0.399013],"study_design_scores_gemma":[0.00004619005,0.00008731658,0.002784541,0.0000293346,0.00005062937,0.00005974622,0.0002998437,0.9862238,0.004767586,0.001589063,0.004024378,0.00003769987],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.7809385,0.0009952721,0.1731526,0.002167953,0.0003990669,0.0004460863,0.006557023,0.02058106,0.01476236],"genre_scores_gemma":[0.9032691,0.0002076743,0.07424189,0.000621437,0.00004045872,0.0001989872,0.01313035,0.000960299,0.007329898],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.6647339,"threshold_uncertainty_score":0.6666292,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02542285309277387,"score_gpt":0.1921105521303376,"score_spread":0.1666876990375637,"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."}}