{"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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.00005246368,0.00006429752,0.00005889603,0.0000327035,0.00007749495,0.0002300105,0.0001187468,0.00001926962,0.000008100652],"category_scores_gemma":[2.901309e-7,0.00004279091,0.00001404081,0.0002782275,0.0000185507,0.000561424,0.00001564615,0.00005315411,0.000005404636],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000004494673,"about_ca_system_score_gemma":0.00001726681,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001638147,"about_ca_topic_score_gemma":0.000007171524,"domain_scores_codex":[0.9995103,0.000007109787,0.00007772692,0.0002179979,0.00008285621,0.0001039425],"domain_scores_gemma":[0.9997786,0.00003116958,0.000008053915,0.0001248167,0.00001701789,0.00004029647],"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":[0.000005106055,0.00001563998,0.000008027891,0.00001862298,0.00001620071,0.000003950296,0.0005030607,0.06776717,0.0003748701,0.8740715,0.001442935,0.05577293],"study_design_scores_gemma":[0.0001048899,0.00002104785,0.00001171893,0.00002410543,0.000005365753,0.00001161822,0.00001108168,0.9915161,0.0001903305,0.003623561,0.004404557,0.00007569572],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.001871106,0.0002626588,0.9876553,0.002383288,0.00004451066,0.0001305162,3.478928e-7,0.0002334942,0.007418708],"genre_scores_gemma":[0.8031203,0.0001879244,0.1945055,0.0003309922,0.00002552274,0.00003196708,0.000002959729,0.000009018635,0.001785832],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9237489,"threshold_uncertainty_score":0.2217997,"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."}}