{"id":"W7123361559","doi":"10.1109/icm66518.2025.11322486","title":"Efficient Pruning and Acceleration of Encoder-Based LLM Transformers on eFPGAs","year":2025,"lang":"","type":"article","venue":"","topic":"Advanced Neural Network Applications","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Encoder; Pruning; Transformer; Computation; Edge device; High memory; Offset (computer science); Decoding methods","routes":{"ca_aff":true,"ca_fund":true,"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":[0.0002124791,0.0002366403,0.0002590456,0.0002486702,0.0003320831,0.00009319035,0.0003947974,0.00009327897,0.00002659894],"category_scores_gemma":[0.00002595035,0.0002275867,0.00007648809,0.001250296,0.0001835353,0.0001278015,0.00007658525,0.0002058296,0.000006369777],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00007524844,"about_ca_system_score_gemma":0.0002049303,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0000109135,"about_ca_topic_score_gemma":0.0000153961,"domain_scores_codex":[0.9981384,0.00006672367,0.0005188518,0.0006427374,0.0002917162,0.0003415164],"domain_scores_gemma":[0.998776,0.0003809235,0.000151738,0.0004654185,0.0001270526,0.00009882791],"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":[0.00005284431,0.0003794913,0.00009303029,0.00009102397,0.00001834493,7.854971e-7,0.0003478948,0.5716984,0.009328547,0.1260792,0.000122067,0.2917883],"study_design_scores_gemma":[0.0006437072,0.0002725453,0.000917572,0.0001644834,0.00002188221,6.509211e-7,0.00004166569,0.9294866,0.06667044,0.0008467915,0.0007599364,0.0001737274],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.05442819,0.0001770141,0.9285731,0.004335818,0.0002007564,0.0007796268,0.000003222467,0.00007153941,0.01143072],"genre_scores_gemma":[0.9666551,0.00004992103,0.03188933,0.000843135,0.00001860461,0.00004602872,0.000002361446,0.000008478512,0.0004870371],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9122269,"threshold_uncertainty_score":0.9280716,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01988389092937311,"score_gpt":0.2851108188255527,"score_spread":0.2652269278961796,"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."}}