{"id":"https://openalex.org/W3206837665","doi":"https://doi.org/10.1109/iccad51958.2021.9643586","title":"Accelerating Framework of Transformer by Hardware Design and Model Compression Co-Optimization","display_name":"Accelerating Framework of Transformer by Hardware Design and Model Compression Co-Optimization","publication_year":2021,"publication_date":"2021-11-01","ids":{"openalex":"https://openalex.org/W3206837665","doi":"https://doi.org/10.1109/iccad51958.2021.9643586","mag":"3206837665"},"language":"en","primary_location":{"id":"doi:10.1109/iccad51958.2021.9643586","is_oa":false,"landing_page_url":"https://doi.org/10.1109/iccad51958.2021.9643586","pdf_url":null,"source":{"id":"https://openalex.org/S4363608354","display_name":"2021 IEEE/ACM International Conference On Computer Aided Design (ICCAD)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2021 IEEE/ACM International Conference On Computer Aided Design (ICCAD)","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":false,"oa_status":"closed","oa_url":null,"any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5077907495","display_name":"Panjie Qi","orcid":null},"institutions":[{"id":"https://openalex.org/I66867065","display_name":"East China Normal University","ror":"https://ror.org/02n96ep67","country_code":"CN","type":"education","lineage":["https://openalex.org/I66867065"]}],"countries":["CN"],"is_corresponding":true,"raw_author_name":"Panjie Qi","raw_affiliation_strings":["East China Normal University, Shanghai, China"],"affiliations":[{"raw_affiliation_string":"East China Normal University, Shanghai, China","institution_ids":["https://openalex.org/I66867065"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5077381252","display_name":"Edwin H.\u2010M. Sha","orcid":"https://orcid.org/0000-0001-5605-5631"},"institutions":[{"id":"https://openalex.org/I66867065","display_name":"East China Normal University","ror":"https://ror.org/02n96ep67","country_code":"CN","type":"education","lineage":["https://openalex.org/I66867065"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Edwin Hsing-Mean Sha","raw_affiliation_strings":["East China Normal University, Shanghai, China"],"affiliations":[{"raw_affiliation_string":"East China Normal University, Shanghai, China","institution_ids":["https://openalex.org/I66867065"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5005358696","display_name":"Qingfeng Zhuge","orcid":"https://orcid.org/0000-0002-1107-3470"},"institutions":[{"id":"https://openalex.org/I66867065","display_name":"East China Normal University","ror":"https://ror.org/02n96ep67","country_code":"CN","type":"education","lineage":["https://openalex.org/I66867065"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Qingfeng Zhuge","raw_affiliation_strings":["East China Normal University, Shanghai, China"],"affiliations":[{"raw_affiliation_string":"East China Normal University, Shanghai, China","institution_ids":["https://openalex.org/I66867065"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5045835605","display_name":"Hongwu Peng","orcid":"https://orcid.org/0000-0003-2025-2195"},"institutions":[{"id":"https://openalex.org/I140172145","display_name":"University of Connecticut","ror":"https://ror.org/02der9h97","country_code":"US","type":"education","lineage":["https://openalex.org/I140172145"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Hongwu Peng","raw_affiliation_strings":["University of Connecticut, Connecticut, USA"],"affiliations":[{"raw_affiliation_string":"University of Connecticut, Connecticut, USA","institution_ids":["https://openalex.org/I140172145"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5073345631","display_name":"Shaoyi Huang","orcid":"https://orcid.org/0000-0001-6093-9798"},"institutions":[{"id":"https://openalex.org/I140172145","display_name":"University of Connecticut","ror":"https://ror.org/02der9h97","country_code":"US","type":"education","lineage":["https://openalex.org/I140172145"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Shaoyi Huang","raw_affiliation_strings":["University of Connecticut, Connecticut, USA"],"affiliations":[{"raw_affiliation_string":"University of Connecticut, Connecticut, USA","institution_ids":["https://openalex.org/I140172145"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5078971265","display_name":"Zhenglun Kong","orcid":"https://orcid.org/0000-0002-8120-4456"},"institutions":[{"id":"https://openalex.org/I12912129","display_name":"Northeastern University","ror":"https://ror.org/04t5xt781","country_code":"US","type":"education","lineage":["https://openalex.org/I12912129"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Zhenglun Kong","raw_affiliation_strings":["Northeastern University, Boston"],"affiliations":[{"raw_affiliation_string":"Northeastern University, Boston","institution_ids":["https://openalex.org/I12912129"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5041623482","display_name":"Yuhong Song","orcid":"https://orcid.org/0000-0002-4310-2766"},"institutions":[{"id":"https://openalex.org/I66867065","display_name":"East China Normal University","ror":"https://ror.org/02n96ep67","country_code":"CN","type":"education","lineage":["https://openalex.org/I66867065"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yuhong Song","raw_affiliation_strings":["East China Normal University, Shanghai, China"],"affiliations":[{"raw_affiliation_string":"East China Normal University, Shanghai, China","institution_ids":["https://openalex.org/I66867065"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5100375952","display_name":"Bingbing Li","orcid":"https://orcid.org/0000-0001-6140-4189"},"institutions":[{"id":"https://openalex.org/I140172145","display_name":"University of Connecticut","ror":"https://ror.org/02der9h97","country_code":"US","type":"education","lineage":["https://openalex.org/I140172145"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Bingbing Li","raw_affiliation_strings":["University of Connecticut, Connecticut, USA"],"affiliations":[{"raw_affiliation_string":"University of Connecticut, Connecticut, USA","institution_ids":["https://openalex.org/I140172145"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":8,"corresponding_author_ids":["https://openalex.org/A5077907495"],"corresponding_institution_ids":["https://openalex.org/I66867065"],"apc_list":null,"apc_paid":null,"fwci":2.2332,"has_fulltext":false,"cited_by_count":47,"citation_normalized_percentile":{"value":0.92252613,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":99,"max":100},"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"9"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10627","display_name":"Advanced Image and Video Retrieval Techniques","score":0.9882000088691711,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T10627","display_name":"Advanced Image and Video Retrieval Techniques","score":0.9882000088691711,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.9868000149726868,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.9796000123023987,"subfield":{"id":"https://openalex.org/subfields/1708","display_name":"Hardware and Architecture"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.7716892957687378},{"id":"https://openalex.org/keywords/transformer","display_name":"Transformer","score":0.650580883026123},{"id":"https://openalex.org/keywords/field-programmable-gate-array","display_name":"Field-programmable gate array","score":0.4872697591781616},{"id":"https://openalex.org/keywords/computer-engineering","display_name":"Computer engineering","score":0.4500378966331482},{"id":"https://openalex.org/keywords/software-deployment","display_name":"Software deployment","score":0.4495835602283478},{"id":"https://openalex.org/keywords/embedded-system","display_name":"Embedded system","score":0.43207767605781555},{"id":"https://openalex.org/keywords/latency","display_name":"Latency (audio)","score":0.423265278339386},{"id":"https://openalex.org/keywords/computer-hardware","display_name":"Computer hardware","score":0.4147956073284149},{"id":"https://openalex.org/keywords/distributed-computing","display_name":"Distributed computing","score":0.3368619680404663},{"id":"https://openalex.org/keywords/electrical-engineering","display_name":"Electrical engineering","score":0.11021250486373901},{"id":"https://openalex.org/keywords/operating-system","display_name":"Operating system","score":0.1044011116027832}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7716892957687378},{"id":"https://openalex.org/C66322947","wikidata":"https://www.wikidata.org/wiki/Q11658","display_name":"Transformer","level":3,"score":0.650580883026123},{"id":"https://openalex.org/C42935608","wikidata":"https://www.wikidata.org/wiki/Q190411","display_name":"Field-programmable gate array","level":2,"score":0.4872697591781616},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.4500378966331482},{"id":"https://openalex.org/C105339364","wikidata":"https://www.wikidata.org/wiki/Q2297740","display_name":"Software deployment","level":2,"score":0.4495835602283478},{"id":"https://openalex.org/C149635348","wikidata":"https://www.wikidata.org/wiki/Q193040","display_name":"Embedded system","level":1,"score":0.43207767605781555},{"id":"https://openalex.org/C82876162","wikidata":"https://www.wikidata.org/wiki/Q17096504","display_name":"Latency (audio)","level":2,"score":0.423265278339386},{"id":"https://openalex.org/C9390403","wikidata":"https://www.wikidata.org/wiki/Q3966","display_name":"Computer hardware","level":1,"score":0.4147956073284149},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.3368619680404663},{"id":"https://openalex.org/C119599485","wikidata":"https://www.wikidata.org/wiki/Q43035","display_name":"Electrical engineering","level":1,"score":0.11021250486373901},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.1044011116027832},{"id":"https://openalex.org/C76155785","wikidata":"https://www.wikidata.org/wiki/Q418","display_name":"Telecommunications","level":1,"score":0.0},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.0},{"id":"https://openalex.org/C165801399","wikidata":"https://www.wikidata.org/wiki/Q25428","display_name":"Voltage","level":2,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/iccad51958.2021.9643586","is_oa":false,"landing_page_url":"https://doi.org/10.1109/iccad51958.2021.9643586","pdf_url":null,"source":{"id":"https://openalex.org/S4363608354","display_name":"2021 IEEE/ACM International Conference On Computer Aided Design (ICCAD)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2021 IEEE/ACM International Conference On Computer Aided Design (ICCAD)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":57,"referenced_works":["https://openalex.org/W1575701986","https://openalex.org/W2009654791","https://openalex.org/W2072806558","https://openalex.org/W2091843288","https://openalex.org/W2114977680","https://openalex.org/W2118463056","https://openalex.org/W2119717200","https://openalex.org/W2511929605","https://openalex.org/W2525332836","https://openalex.org/W2553303224","https://openalex.org/W2767785892","https://openalex.org/W2794952988","https://openalex.org/W2796625795","https://openalex.org/W2805249667","https://openalex.org/W2896457183","https://openalex.org/W2911884654","https://openalex.org/W2912521296","https://openalex.org/W2915106038","https://openalex.org/W2917450576","https://openalex.org/W2923014074","https://openalex.org/W2942063194","https://openalex.org/W2951008357","https://openalex.org/W2962953210","https://openalex.org/W2962958286","https://openalex.org/W2963310665","https://openalex.org/W2963341956","https://openalex.org/W2963403868","https://openalex.org/W2964045208","https://openalex.org/W2978017171","https://openalex.org/W2980856918","https://openalex.org/W2998342322","https://openalex.org/W3017024317","https://openalex.org/W3025085339","https://openalex.org/W3035251378","https://openalex.org/W3047848469","https://openalex.org/W3091170309","https://openalex.org/W3092334294","https://openalex.org/W3100980998","https://openalex.org/W3103754749","https://openalex.org/W3104263050","https://openalex.org/W3104401858","https://openalex.org/W3121413886","https://openalex.org/W3132616766","https://openalex.org/W4230252700","https://openalex.org/W4298422451","https://openalex.org/W4385245566","https://openalex.org/W6638318767","https://openalex.org/W6727099177","https://openalex.org/W6729956949","https://openalex.org/W6739901393","https://openalex.org/W6746331415","https://openalex.org/W6752273154","https://openalex.org/W6755207826","https://openalex.org/W6758657797","https://openalex.org/W6762510946","https://openalex.org/W6768851824","https://openalex.org/W6780482815"],"related_works":["https://openalex.org/W96612179","https://openalex.org/W2770234245","https://openalex.org/W2566006169","https://openalex.org/W2987774938","https://openalex.org/W632915154","https://openalex.org/W4229499248","https://openalex.org/W4378874356","https://openalex.org/W2055733372","https://openalex.org/W2369811061","https://openalex.org/W3089997100"],"abstract_inverted_index":{"State-of-the-art":[0],"Transformer-based":[1],"models,":[2],"with":[3,17,38,50,147],"gigantic":[4],"parameters,":[5],"are":[6,27],"difficult":[7],"to":[8,29,75,88,102,142,170,187,218,247],"be":[9,46],"accommodated":[10],"on":[11,196],"resource":[12],"constrained":[13],"embedded":[14,25],"devices.":[15,67,220],"Moreover,":[16],"the":[18,62,82,94,100,104,162,189],"development":[19],"of":[20,90,97,185,191],"technology,":[21],"more":[22,24,230],"and":[23,85,99,126,198,212,228,250],"devices":[26,49,208,217],"available":[28],"run":[30],"a":[31,35,119,121,134,144,153,180],"Transformer":[32,36,98,197],"model.":[33],"For":[34],"model":[37,146,200],"different":[39,51,207],"constraints":[40],"(tight":[41],"or":[42],"loose),":[43],"it":[44],"can":[45,132,205,223,238],"deployed":[47],"onto":[48],"computing":[52],"power.":[53],"However,":[54],"in":[55],"previous":[56],"work,":[57],"designers":[58],"did":[59],"not":[60,80],"choose":[61],"best":[63,83,105,135],"device":[64,74,136],"among":[65],"multiple":[66],"Instead,":[68],"they":[69],"just":[70],"used":[71],"an":[72,109],"existing":[73],"deploy":[76],"model,":[77,122],"which":[78],"was":[79],"necessarily":[81],"fit":[84],"may":[86],"lead":[87],"underutilization":[89],"resources.":[91],"To":[92],"address":[93],"deployment":[95],"challenge":[96],"problem":[101,190],"select":[103],"device,":[106],"we":[107,151],"propose":[108,152],"algorithm":[110],"<tex":[111],"xmlns:mml=\"http://www.w3.org/1998/Math/MathML\"":[112],"xmlns:xlink=\"http://www.w3.org/1999/xlink\">$\\leftrightarrows$</tex>":[113],"hardware":[114],"closed-loop":[115],"acceleration":[116],"framework.":[117],"Given":[118],"dataset,":[120],"latency":[123],"constraint":[124,128],"LC":[125,211],"accuracy":[127],"AC,":[129,213],"our":[130,203],"framework":[131,204,237],"provide":[133],"satisfying":[137],"both":[138],"constraints.":[139],"In":[140],"order":[141],"generate":[143],"compressed":[145],"high":[148],"sparsity":[149,226,234],"ratio,":[150],"novel":[154],"pruning":[155,158],"technique,":[156],"hierarchical":[157],"(HP).":[159],"We":[160,178],"optimize":[161],"sparse":[163],"matrix":[164,169],"storage":[165],"format":[166],"for":[167,175,209],"HP":[168,186,222],"further":[171],"reduce":[172],"memory":[173],"usage":[174],"FPGA":[176],"implementation.":[177],"design":[179],"accelerator":[181],"that":[182,202],"takes":[183],"advantage":[184],"solve":[188],"concurrent":[192],"random":[193],"access.":[194],"Experiments":[195],"TinyBert":[199],"show":[201],"find":[206],"various":[210],"covering":[214],"from":[215],"low-end":[216],"high-end":[219],"Our":[221,236],"achieve":[224,239],"higher":[225],"ratio":[227],"is":[229],"flexible":[231],"than":[232],"other":[233],"pattern.":[235],"37":[240],"x,":[241,243],"1.9":[242],"1.7x":[244],"speedup":[245],"compared":[246],"CPU,":[248],"GPU":[249],"FPGA,":[251],"respectively.":[252]},"counts_by_year":[{"year":2026,"cited_by_count":3},{"year":2025,"cited_by_count":10},{"year":2024,"cited_by_count":15},{"year":2023,"cited_by_count":9},{"year":2022,"cited_by_count":10}],"updated_date":"2026-04-03T22:45:19.894376","created_date":"2025-10-10T00:00:00"}
