{"id":"W4389518890","doi":"10.18653/v1/2023.findings-emnlp.634","title":"Mixture-of-Linguistic-Experts Adapters for Improving and Interpreting Pre-trained Language Models","year":2023,"lang":"en","type":"article","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of the Fraser Valley; University of British Columbia","funders":"","keywords":"Softmax function; Computer science; Language model; Gumbel distribution; Adapter (computing); Layer (electronics); Artificial intelligence; Pruning; Natural language processing; Encoding (memory); Artificial neural network; Mathematics; Statistics; Extreme value theory","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":[0.0003010206,0.0000973534,0.0001361049,0.00009364888,0.00005280499,0.00006207309,0.0003349074,0.00004175523,0.000001887033],"category_scores_gemma":[0.00024058,0.00008554808,0.00004402472,0.0001182752,0.00001859865,0.000167774,0.0002585084,0.00004311344,5.492353e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0000125143,"about_ca_system_score_gemma":0.00002579997,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001164301,"about_ca_topic_score_gemma":0.00001831068,"domain_scores_codex":[0.9991117,0.00001526777,0.0002120748,0.0003154334,0.0001183049,0.0002272843],"domain_scores_gemma":[0.9993336,0.000207556,0.00005975925,0.0002939891,0.00005068711,0.00005443773],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00004954531,0.00004075661,0.0001979089,0.0005624923,0.00005878367,0.00003459725,0.1554835,0.009129913,0.0709974,0.2492623,0.0004658534,0.5137169],"study_design_scores_gemma":[0.0001763817,0.00003299113,0.00001087815,0.00003450601,0.000003425539,0.000002647261,0.0004052261,0.9905822,0.002583867,0.006027017,0.0000381905,0.000102618],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.09519447,0.0000749621,0.9029168,0.0001943913,0.0002127821,0.0001862093,0.000002102378,0.0002642366,0.0009540677],"genre_scores_gemma":[0.8026399,0.000001494829,0.196885,0.0001197696,0.00004949706,0.00001633399,0.000001404588,0.000008392498,0.0002782222],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9814523,"threshold_uncertainty_score":0.348855,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02013140030204983,"score_gpt":0.2764942223730839,"score_spread":0.2563628220710341,"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."}}