{"id":"https://openalex.org/W4390356856","doi":"https://doi.org/10.1109/tcad.2023.3347291","title":"Mobile Transformer Accelerator Exploiting Various Line Sparsity and Tile-Based Dynamic Quantization","display_name":"Mobile Transformer Accelerator Exploiting Various Line Sparsity and Tile-Based Dynamic Quantization","publication_year":2023,"publication_date":"2023-12-28","ids":{"openalex":"https://openalex.org/W4390356856","doi":"https://doi.org/10.1109/tcad.2023.3347291"},"language":"en","primary_location":{"id":"doi:10.1109/tcad.2023.3347291","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tcad.2023.3347291","pdf_url":null,"source":{"id":"https://openalex.org/S100835903","display_name":"IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems","issn_l":"0278-0070","issn":["0278-0070","1937-4151"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319808","host_organization_name":"Institute of Electrical and Electronics Engineers","host_organization_lineage":["https://openalex.org/P4310319808"],"host_organization_lineage_names":["Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems","raw_type":"journal-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/A5035447590","display_name":"Eunji Kwon","orcid":"https://orcid.org/0000-0001-5313-8976"},"institutions":[{"id":"https://openalex.org/I123900574","display_name":"Pohang University of Science and Technology","ror":"https://ror.org/04xysgw12","country_code":"KR","type":"education","lineage":["https://openalex.org/I123900574"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Eunji Kwon","raw_affiliation_strings":["Electrical Engineering Department, Pohang University of Science and Technology, Pohang, South Korea"],"raw_orcid":"https://orcid.org/0000-0001-5313-8976","affiliations":[{"raw_affiliation_string":"Electrical Engineering Department, Pohang University of Science and Technology, Pohang, South Korea","institution_ids":["https://openalex.org/I123900574"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5017591432","display_name":"Jongho Yoon","orcid":"https://orcid.org/0000-0002-9295-4850"},"institutions":[{"id":"https://openalex.org/I123900574","display_name":"Pohang University of Science and Technology","ror":"https://ror.org/04xysgw12","country_code":"KR","type":"education","lineage":["https://openalex.org/I123900574"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Jongho Yoon","raw_affiliation_strings":["Electrical Engineering Department, Pohang University of Science and Technology, Pohang, South Korea"],"raw_orcid":"https://orcid.org/0000-0002-9295-4850","affiliations":[{"raw_affiliation_string":"Electrical Engineering Department, Pohang University of Science and Technology, Pohang, South Korea","institution_ids":["https://openalex.org/I123900574"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5044543389","display_name":"Seokhyeong Kang","orcid":"https://orcid.org/0000-0003-3015-1806"},"institutions":[{"id":"https://openalex.org/I123900574","display_name":"Pohang University of Science and Technology","ror":"https://ror.org/04xysgw12","country_code":"KR","type":"education","lineage":["https://openalex.org/I123900574"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Seokhyeong Kang","raw_affiliation_strings":["Electrical Engineering Department, Pohang University of Science and Technology, Pohang, South Korea"],"raw_orcid":"https://orcid.org/0000-0003-3015-1806","affiliations":[{"raw_affiliation_string":"Electrical Engineering Department, Pohang University of Science and Technology, Pohang, South Korea","institution_ids":["https://openalex.org/I123900574"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I123900574"],"apc_list":null,"apc_paid":null,"fwci":2.7636,"has_fulltext":false,"cited_by_count":12,"citation_normalized_percentile":{"value":0.90071648,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":97,"max":99},"biblio":{"volume":"43","issue":"6","first_page":"1808","last_page":"1821"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.972100019454956,"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"}},"topics":[{"id":"https://openalex.org/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.972100019454956,"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"}},{"id":"https://openalex.org/T12611","display_name":"Neural Networks and Reservoir Computing","score":0.939300000667572,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"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/T10792","display_name":"Matrix Theory and Algorithms","score":0.9103999733924866,"subfield":{"id":"https://openalex.org/subfields/1703","display_name":"Computational Theory and Mathematics"},"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":0.7764379382133484},{"id":"https://openalex.org/keywords/transformer","display_name":"Transformer","score":0.633538007736206},{"id":"https://openalex.org/keywords/efficient-energy-use","display_name":"Efficient energy use","score":0.53407883644104},{"id":"https://openalex.org/keywords/computation","display_name":"Computation","score":0.4771418273448944},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.4768087565898895},{"id":"https://openalex.org/keywords/quantization","display_name":"Quantization (signal processing)","score":0.4719851613044739},{"id":"https://openalex.org/keywords/matrix-multiplication","display_name":"Matrix multiplication","score":0.4231621325016022},{"id":"https://openalex.org/keywords/mobile-device","display_name":"Mobile device","score":0.4222533702850342},{"id":"https://openalex.org/keywords/granularity","display_name":"Granularity","score":0.41283875703811646},{"id":"https://openalex.org/keywords/computer-hardware","display_name":"Computer hardware","score":0.3911658525466919},{"id":"https://openalex.org/keywords/computational-science","display_name":"Computational science","score":0.3488028347492218},{"id":"https://openalex.org/keywords/computer-engineering","display_name":"Computer engineering","score":0.34544461965560913},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.20947867631912231},{"id":"https://openalex.org/keywords/electrical-engineering","display_name":"Electrical engineering","score":0.15080636739730835},{"id":"https://openalex.org/keywords/voltage","display_name":"Voltage","score":0.11644437909126282}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7764379382133484},{"id":"https://openalex.org/C66322947","wikidata":"https://www.wikidata.org/wiki/Q11658","display_name":"Transformer","level":3,"score":0.633538007736206},{"id":"https://openalex.org/C2742236","wikidata":"https://www.wikidata.org/wiki/Q924713","display_name":"Efficient energy use","level":2,"score":0.53407883644104},{"id":"https://openalex.org/C45374587","wikidata":"https://www.wikidata.org/wiki/Q12525525","display_name":"Computation","level":2,"score":0.4771418273448944},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.4768087565898895},{"id":"https://openalex.org/C28855332","wikidata":"https://www.wikidata.org/wiki/Q198099","display_name":"Quantization (signal processing)","level":2,"score":0.4719851613044739},{"id":"https://openalex.org/C17349429","wikidata":"https://www.wikidata.org/wiki/Q1049914","display_name":"Matrix multiplication","level":3,"score":0.4231621325016022},{"id":"https://openalex.org/C186967261","wikidata":"https://www.wikidata.org/wiki/Q5082128","display_name":"Mobile device","level":2,"score":0.4222533702850342},{"id":"https://openalex.org/C177774035","wikidata":"https://www.wikidata.org/wiki/Q1246948","display_name":"Granularity","level":2,"score":0.41283875703811646},{"id":"https://openalex.org/C9390403","wikidata":"https://www.wikidata.org/wiki/Q3966","display_name":"Computer hardware","level":1,"score":0.3911658525466919},{"id":"https://openalex.org/C459310","wikidata":"https://www.wikidata.org/wiki/Q117801","display_name":"Computational science","level":1,"score":0.3488028347492218},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.34544461965560913},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.20947867631912231},{"id":"https://openalex.org/C119599485","wikidata":"https://www.wikidata.org/wiki/Q43035","display_name":"Electrical engineering","level":1,"score":0.15080636739730835},{"id":"https://openalex.org/C165801399","wikidata":"https://www.wikidata.org/wiki/Q25428","display_name":"Voltage","level":2,"score":0.11644437909126282},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.0},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.0},{"id":"https://openalex.org/C84114770","wikidata":"https://www.wikidata.org/wiki/Q46344","display_name":"Quantum","level":2,"score":0.0},{"id":"https://openalex.org/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","level":0,"score":0.0},{"id":"https://openalex.org/C62520636","wikidata":"https://www.wikidata.org/wiki/Q944","display_name":"Quantum mechanics","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/tcad.2023.3347291","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tcad.2023.3347291","pdf_url":null,"source":{"id":"https://openalex.org/S100835903","display_name":"IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems","issn_l":"0278-0070","issn":["0278-0070","1937-4151"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319808","host_organization_name":"Institute of Electrical and Electronics Engineers","host_organization_lineage":["https://openalex.org/P4310319808"],"host_organization_lineage_names":["Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems","raw_type":"journal-article"}],"best_oa_location":null,"sustainable_development_goals":[{"score":0.9100000262260437,"display_name":"Affordable and clean energy","id":"https://metadata.un.org/sdg/7"}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":39,"referenced_works":["https://openalex.org/W639708223","https://openalex.org/W2094756095","https://openalex.org/W2896409484","https://openalex.org/W2921918777","https://openalex.org/W2946794439","https://openalex.org/W2975381464","https://openalex.org/W2986915867","https://openalex.org/W3017024317","https://openalex.org/W3035251378","https://openalex.org/W3047848469","https://openalex.org/W3094502228","https://openalex.org/W3096609285","https://openalex.org/W3105802176","https://openalex.org/W3154851733","https://openalex.org/W3159727696","https://openalex.org/W3162542754","https://openalex.org/W3194959296","https://openalex.org/W3196923642","https://openalex.org/W3206837665","https://openalex.org/W4211085526","https://openalex.org/W4221144813","https://openalex.org/W4287116734","https://openalex.org/W4288791664","https://openalex.org/W4293025165","https://openalex.org/W4308479898","https://openalex.org/W4312950730","https://openalex.org/W4385245566","https://openalex.org/W6755843862","https://openalex.org/W6762945437","https://openalex.org/W6768080748","https://openalex.org/W6788001715","https://openalex.org/W6790690058","https://openalex.org/W6796931752","https://openalex.org/W6798046796","https://openalex.org/W6798160016","https://openalex.org/W6800633732","https://openalex.org/W6810005255","https://openalex.org/W6840518963","https://openalex.org/W6847478871"],"related_works":["https://openalex.org/W2931688134","https://openalex.org/W2377919138","https://openalex.org/W2378857091","https://openalex.org/W4256502920","https://openalex.org/W103652678","https://openalex.org/W3183118997","https://openalex.org/W3214410901","https://openalex.org/W3204400881","https://openalex.org/W3204296682","https://openalex.org/W2917767146"],"abstract_inverted_index":{"Transformer":[0],"models":[1],"are":[2,47],"difficult":[3],"to":[4,10,50,160],"employ":[5],"in":[6,95,131],"mobile":[7,161,196],"devices":[8],"due":[9],"their":[11],"memory-and":[12],"computation-intensive":[13],"properties.":[14],"Accordingly,":[15],"there":[16],"is":[17],"ongoing":[18],"research":[19],"on":[20],"various":[21,78],"methods":[22],"for":[23,73,176,185],"compressing":[24],"transformer":[25,179,188,208],"models,":[26,190],"such":[27,36],"as":[28,37],"pruning":[29,86],"and":[30,42,98,106,114,149,163,174,181,183],"quantization.":[31],"However,":[32],"general":[33,140],"computing":[34],"platforms":[35],"central":[38],"processing":[39,44,121],"units":[40,45],"(CPUs)":[41],"graphics":[43],"(GPUs)":[46],"not":[48],"energy-efficient":[49],"accelerate":[51],"the":[52,56,116,126,165,169,177,186,199,204],"pruned":[53,93],"model":[54],"because":[55],"unstructured":[57],"sparsity":[58,82],"they":[59],"exhibit":[60],"causes":[61],"degradation":[62],"of":[63,80,129,154],"parallelism.":[64],"In":[65,192],"this":[66],"paper,":[67],"we":[68,135],"propose":[69],"a":[70,96,103,137,157],"low-power":[71],"accelerator":[72,167,197],"transformers":[74,94],"that":[75,109,145],"can":[76],"handle":[77],"levels":[79],"structured":[81],"induced":[83],"by":[84,125,172],"line":[85],"with":[87],"different":[88],"granularity.":[89],"Our":[90],"approach":[91],"accelerates":[92],"head-wise":[97,111],"line-wise":[99,147],"manner.":[100],"We":[101],"present":[102],"head":[104],"reorganization":[105],"shuffling":[107],"method":[108],"supports":[110,146],"skip":[112],"operations":[113,130],"resolves":[115],"load":[117],"imbalance":[118],"problem":[119],"among":[120,203],"engines":[122],"(PEs)":[123],"caused":[124],"varying":[127],"number":[128],"each":[132],"head.":[133],"Furthermore,":[134],"implemented":[136],"sparse":[138],"quantized":[139],"matrix-to-matrix":[141],"multiplication":[142],"(SQ-GEMM)":[143],"module":[144],"skipping":[148],"on-the-fly":[150],"tile-based":[151],"dynamic":[152],"quantization":[153],"activations.":[155],"As":[156],"result,":[158],"compared":[159],"GPU":[162],"CPU,":[164],"proposed":[166,195],"improved":[168],"energy":[170,201],"efficiency":[171,202],"2.9\u00d7":[173],"12.3\u00d7":[175],"detection":[178],"(DETR),":[180],"3.0\u00d7":[182],"12.4\u00d7":[184],"vision":[187],"(ViT)":[189],"respectively.":[191],"addition,":[193],"our":[194],"achieved":[198],"highest":[200],"current":[205],"state-of-the-art":[206],"FPGA-based":[207],"accelerators.":[209]},"counts_by_year":[{"year":2026,"cited_by_count":2},{"year":2025,"cited_by_count":6},{"year":2024,"cited_by_count":4}],"updated_date":"2026-08-08T01:25:22.217667","created_date":"2025-10-10T00:00:00"}