{"id":"W4400230261","doi":"10.1109/iscas58744.2024.10558045","title":"Optimized Transformer Models: ℓ′ BERT with CNN-like Pruning and Quantization","year":2024,"lang":"en","type":"article","venue":"","topic":"Neural Networks and Applications","field":"Computer Science","cited_by":3,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Computer science; Transformer; Quantization (signal processing); Pruning; Artificial intelligence; Algorithm; Electrical engineering; Engineering; Voltage","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"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":[0.0004358676,0.0006338067,0.0005670594,0.0003381322,0.0002037199,0.0007648869,0.00120661,0.0006377067,0.003111769],"category_scores_gemma":[0.001886277,0.0002838073,0.0004973272,0.000371605,0.0003617491,0.001256962,0.0006137716,0.0009334018,0.000652462],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0008891137,"about_ca_system_score_gemma":0.0008912299,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00690307,"about_ca_topic_score_gemma":0.01098846,"domain_scores_codex":[0.9998726,0.00002186636,0.000008387868,0.00002802713,0.00004630778,0.00002280972],"domain_scores_gemma":[0.9996444,0.0001607673,0.00003482245,0.00005229762,0.00008712053,0.00002057022],"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":[0.0001468136,0.00004753661,0.0006842423,0.00006848184,0.00003164245,0.00008255037,0.00003842941,0.8269079,0.004753171,0.02033974,0.003439253,0.1434602],"study_design_scores_gemma":[0.000003280679,0.0000121739,0.00002686041,0.000003558814,0.00000363244,0.00001104428,0.00000264156,0.996725,0.0007637109,0.002090926,0.0003550878,0.000002097519],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.04014743,0.0005660151,0.9496385,0.0002625258,0.00009632719,0.00006143324,0.0001790241,0.001082194,0.00796658],"genre_scores_gemma":[0.7822593,0.0005381716,0.2057684,0.0002876845,0.0000491654,0.0001219877,0.0004668779,0.0003060861,0.01020233],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.00690307,"threshold_uncertainty_score":0.01372582,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01689673961901262,"score_gpt":0.2270784856665521,"score_spread":0.2101817460475394,"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."}}