{"id":"https://openalex.org/W7152750420","doi":"https://doi.org/10.1007/s11227-026-08450-4","title":"An embedded vision transformer with quantization-friendly simple attention, optimized buffers, and SIMD acceleration","display_name":"An embedded vision transformer with quantization-friendly simple attention, optimized buffers, and SIMD acceleration","publication_year":2026,"publication_date":"2026-04-09","ids":{"openalex":"https://openalex.org/W7152750420","doi":"https://doi.org/10.1007/s11227-026-08450-4"},"language":"en","primary_location":{"id":"doi:10.1007/s11227-026-08450-4","is_oa":true,"landing_page_url":"https://doi.org/10.1007/s11227-026-08450-4","pdf_url":"https://link.springer.com/content/pdf/10.1007/s11227-026-08450-4.pdf","source":{"id":"https://openalex.org/S32326811","display_name":"The Journal of Supercomputing","issn_l":"0920-8542","issn":["0920-8542","1573-0484"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319900","host_organization_name":"Springer Science+Business Media","host_organization_lineage":["https://openalex.org/P4310319900","https://openalex.org/P4310319965"],"host_organization_lineage_names":["Springer Science+Business Media","Springer Nature"],"type":"journal"},"license":"cc-by-nc-nd","license_id":"https://openalex.org/licenses/cc-by-nc-nd","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"The Journal of Supercomputing","raw_type":"journal-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"hybrid","oa_url":"https://link.springer.com/content/pdf/10.1007/s11227-026-08450-4.pdf","any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5050351696","display_name":"Hayata Kaneko","orcid":null},"institutions":[{"id":"https://openalex.org/I135768898","display_name":"Ritsumeikan University","ror":"https://ror.org/0197nmd03","country_code":"JP","type":"education","lineage":["https://openalex.org/I135768898","https://openalex.org/I4390039241"]}],"countries":["JP"],"is_corresponding":false,"raw_author_name":"Hayata Kaneko","raw_affiliation_strings":["Department of Electronic and Computer Engineering, Ritsumeikan University, Noji-higashi, Kusatsu, Shiga, 525-8577, Japan"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Department of Electronic and Computer Engineering, Ritsumeikan University, Noji-higashi, Kusatsu, Shiga, 525-8577, Japan","institution_ids":["https://openalex.org/I135768898"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5133277434","display_name":"Lin Meng","orcid":null},"institutions":[{"id":"https://openalex.org/I135768898","display_name":"Ritsumeikan University","ror":"https://ror.org/0197nmd03","country_code":"JP","type":"education","lineage":["https://openalex.org/I135768898","https://openalex.org/I4390039241"]}],"countries":["JP"],"is_corresponding":true,"raw_author_name":"Lin Meng","raw_affiliation_strings":["Department of Electronic and Computer Engineering, Ritsumeikan University, Noji-higashi, Kusatsu, Shiga, 525-8577, Japan"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Department of Electronic and Computer Engineering, Ritsumeikan University, Noji-higashi, Kusatsu, Shiga, 525-8577, Japan","institution_ids":["https://openalex.org/I135768898"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":["https://openalex.org/A5133277434"],"corresponding_institution_ids":["https://openalex.org/I135768898"],"apc_list":{"value":2990,"currency":"USD","value_usd":2990},"apc_paid":{"value":2990,"currency":"USD","value_usd":2990},"fwci":0.0,"has_fulltext":true,"cited_by_count":0,"citation_normalized_percentile":{"value":0.44137674,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":"82","issue":"5","first_page":null,"last_page":null},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.3637000024318695,"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/T10036","display_name":"Advanced Neural Network Applications","score":0.3637000024318695,"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/T11992","display_name":"CCD and CMOS Imaging Sensors","score":0.21289999783039093,"subfield":{"id":"https://openalex.org/subfields/2208","display_name":"Electrical and Electronic Engineering"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10901","display_name":"Advanced Data Compression Techniques","score":0.09489999711513519,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/simd","display_name":"SIMD","score":0.7656000256538391},{"id":"https://openalex.org/keywords/quantization","display_name":"Quantization (signal processing)","score":0.595300018787384},{"id":"https://openalex.org/keywords/memory-footprint","display_name":"Memory footprint","score":0.45669999718666077},{"id":"https://openalex.org/keywords/normalization","display_name":"Normalization (sociology)","score":0.4262000024318695},{"id":"https://openalex.org/keywords/hardware-acceleration","display_name":"Hardware acceleration","score":0.3849000036716461},{"id":"https://openalex.org/keywords/arithmetic-coding","display_name":"Arithmetic coding","score":0.35280001163482666},{"id":"https://openalex.org/keywords/transformer","display_name":"Transformer","score":0.328900009393692},{"id":"https://openalex.org/keywords/integer-programming","display_name":"Integer programming","score":0.325300008058548}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8517000079154968},{"id":"https://openalex.org/C150552126","wikidata":"https://www.wikidata.org/wiki/Q339387","display_name":"SIMD","level":2,"score":0.7656000256538391},{"id":"https://openalex.org/C28855332","wikidata":"https://www.wikidata.org/wiki/Q198099","display_name":"Quantization (signal processing)","level":2,"score":0.595300018787384},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.5005999803543091},{"id":"https://openalex.org/C74912251","wikidata":"https://www.wikidata.org/wiki/Q6815727","display_name":"Memory footprint","level":2,"score":0.45669999718666077},{"id":"https://openalex.org/C136886441","wikidata":"https://www.wikidata.org/wiki/Q926129","display_name":"Normalization (sociology)","level":2,"score":0.4262000024318695},{"id":"https://openalex.org/C13164978","wikidata":"https://www.wikidata.org/wiki/Q600158","display_name":"Hardware acceleration","level":3,"score":0.3849000036716461},{"id":"https://openalex.org/C9390403","wikidata":"https://www.wikidata.org/wiki/Q3966","display_name":"Computer hardware","level":1,"score":0.37959998846054077},{"id":"https://openalex.org/C153338461","wikidata":"https://www.wikidata.org/wiki/Q2651","display_name":"Arithmetic coding","level":4,"score":0.35280001163482666},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.3441999852657318},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.3422999978065491},{"id":"https://openalex.org/C66322947","wikidata":"https://www.wikidata.org/wiki/Q11658","display_name":"Transformer","level":3,"score":0.328900009393692},{"id":"https://openalex.org/C149635348","wikidata":"https://www.wikidata.org/wiki/Q193040","display_name":"Embedded system","level":1,"score":0.32589998841285706},{"id":"https://openalex.org/C56086750","wikidata":"https://www.wikidata.org/wiki/Q6042592","display_name":"Integer programming","level":2,"score":0.325300008058548},{"id":"https://openalex.org/C2780728072","wikidata":"https://www.wikidata.org/wiki/Q5297","display_name":"Microprocessor","level":2,"score":0.30160000920295715},{"id":"https://openalex.org/C40842320","wikidata":"https://www.wikidata.org/wiki/Q19423","display_name":"Buffer overflow","level":2,"score":0.2741999924182892},{"id":"https://openalex.org/C88548561","wikidata":"https://www.wikidata.org/wiki/Q347599","display_name":"sort","level":2,"score":0.26930001378059387},{"id":"https://openalex.org/C42935608","wikidata":"https://www.wikidata.org/wiki/Q190411","display_name":"Field-programmable gate array","level":2,"score":0.2639999985694885},{"id":"https://openalex.org/C45374587","wikidata":"https://www.wikidata.org/wiki/Q12525525","display_name":"Computation","level":2,"score":0.26170000433921814},{"id":"https://openalex.org/C179518139","wikidata":"https://www.wikidata.org/wiki/Q5140297","display_name":"Coding (social sciences)","level":2,"score":0.26170000433921814},{"id":"https://openalex.org/C82876162","wikidata":"https://www.wikidata.org/wiki/Q17096504","display_name":"Latency (audio)","level":2,"score":0.2612000107765198},{"id":"https://openalex.org/C97137487","wikidata":"https://www.wikidata.org/wiki/Q729138","display_name":"Integer (computer science)","level":2,"score":0.2596000134944916},{"id":"https://openalex.org/C153083717","wikidata":"https://www.wikidata.org/wiki/Q6535263","display_name":"Leverage (statistics)","level":2,"score":0.2547999918460846},{"id":"https://openalex.org/C26713055","wikidata":"https://www.wikidata.org/wiki/Q245962","display_name":"Implementation","level":2,"score":0.2517000138759613}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1007/s11227-026-08450-4","is_oa":true,"landing_page_url":"https://doi.org/10.1007/s11227-026-08450-4","pdf_url":"https://link.springer.com/content/pdf/10.1007/s11227-026-08450-4.pdf","source":{"id":"https://openalex.org/S32326811","display_name":"The Journal of Supercomputing","issn_l":"0920-8542","issn":["0920-8542","1573-0484"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319900","host_organization_name":"Springer Science+Business Media","host_organization_lineage":["https://openalex.org/P4310319900","https://openalex.org/P4310319965"],"host_organization_lineage_names":["Springer Science+Business Media","Springer Nature"],"type":"journal"},"license":"cc-by-nc-nd","license_id":"https://openalex.org/licenses/cc-by-nc-nd","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"The Journal of Supercomputing","raw_type":"journal-article"}],"best_oa_location":{"id":"doi:10.1007/s11227-026-08450-4","is_oa":true,"landing_page_url":"https://doi.org/10.1007/s11227-026-08450-4","pdf_url":"https://link.springer.com/content/pdf/10.1007/s11227-026-08450-4.pdf","source":{"id":"https://openalex.org/S32326811","display_name":"The Journal of Supercomputing","issn_l":"0920-8542","issn":["0920-8542","1573-0484"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319900","host_organization_name":"Springer Science+Business Media","host_organization_lineage":["https://openalex.org/P4310319900","https://openalex.org/P4310319965"],"host_organization_lineage_names":["Springer Science+Business Media","Springer Nature"],"type":"journal"},"license":"cc-by-nc-nd","license_id":"https://openalex.org/licenses/cc-by-nc-nd","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"The Journal of Supercomputing","raw_type":"journal-article"},"sustainable_development_goals":[],"awards":[],"funders":[{"id":"https://openalex.org/F4320323289","display_name":"Ritsumeikan University","ror":"https://ror.org/0197nmd03"},{"id":"https://openalex.org/F4320334764","display_name":"Japan Society for the Promotion of Science","ror":"https://ror.org/00hhkn466"}],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7152750420.pdf","grobid_xml":"https://content.openalex.org/works/W7152750420.grobid-xml"},"referenced_works_count":22,"referenced_works":["https://openalex.org/W1861492603","https://openalex.org/W2889339831","https://openalex.org/W2963122961","https://openalex.org/W2963255460","https://openalex.org/W4285601701","https://openalex.org/W4306955484","https://openalex.org/W4312885563","https://openalex.org/W4312950730","https://openalex.org/W4319430529","https://openalex.org/W4362683541","https://openalex.org/W4386322953","https://openalex.org/W4390873361","https://openalex.org/W4391331710","https://openalex.org/W4394625669","https://openalex.org/W4398757502","https://openalex.org/W4400877686","https://openalex.org/W4403918843","https://openalex.org/W4405926999","https://openalex.org/W4413978821","https://openalex.org/W4415207016","https://openalex.org/W4417166139","https://openalex.org/W7133227460"],"related_works":[],"abstract_inverted_index":{"Vision":[0],"Transformer":[1],"(ViT)":[2],"is":[3],"the":[4,15,134,172,186,192],"key":[5],"to":[6,14,71,113,125],"state-of-the-art":[7],"DNN":[8],"models":[9],"in":[10,147,178],"computer":[11],"vision.":[12],"Due":[13],"high":[16],"memory":[17,78,149],"requirement":[18],"and":[19,27,42,64,80,118,143,160,191],"computational":[20],"burden":[21],"of":[22,174,188],"ViTs,":[23],"a":[24,53,90,132,151],"lightweight":[25,36],"design":[26],"integer":[28,92],"quantization":[29,190],"are":[30],"essential":[31],"approaches":[32],".":[33],"However,":[34],"these":[35],"implementations":[37],"still":[38],"target":[39],"desktop-class":[40],"hardware":[41],"high-end":[43],"edge":[44],"devices.":[45],"The":[46],"deployment":[47],"on":[48,150],"resource-restricted":[49],"embedded":[50,73,104],"devices":[51],"remains":[52],"significant":[54],"challenge.":[55],"In":[56],"this":[57],"context,":[58],"floating-point":[59],"arithmetic":[60],"avoidance,":[61],"buffer":[62,119,124,128],"management,":[63],"CPU\u2019s":[65],"SIMD":[66,158],"leverage":[67],"cannot":[68],"be":[69,138,145],"ignored":[70],"meet":[72],"system":[74],"requirements,":[75],"such":[76],"as":[77],"footprint":[79],"real-time":[81],"execution.":[82],"Therefore,":[83],"we":[84,107],"implement":[85],"an":[86],"ultra-lightweight":[87],"ViT,":[88],"i.e.,":[89],"fully":[91],"CNN-Transformer":[93],"hybrid":[94],"model":[95],"with":[96,157,185],"Simple":[97,175],"attention":[98],"by":[99],"Post-Training":[100],"Quantization":[101,111],"(PTQ),":[102],"targeting":[103],"hardware.":[105],"Specifically,":[106],"introduce":[108],"Normalization":[109],"Equivalent":[110],"(NEQ)":[112],"reduce":[114],"l-1":[115],"normalization":[116],"computation,":[117],"optimization,":[120],"e.g.,":[121],"Attention":[122],"row":[123],"mitigate":[126],"$$O(n^2)$$":[127],"size":[129,136],"increase.":[130],"As":[131],"result,":[133],"parameter":[135],"can":[137,144],"kept":[139],"under":[140],"512":[141],"KiB":[142],"deployed":[146],"on-chip":[148],"small-scale":[152],"RISC-V":[153],"core,":[154],"CV32E40p.":[155],"Additionally,":[156],"extension":[159],"its":[161],"dedicated":[162],"kernels,":[163],"our":[164],"implementation":[165],"achieves":[166],"10.4":[167],"$$\\times":[168,180],"$$":[169,181],"acceleration":[170],"for":[171],"logit":[173],"attention,":[176],"resulting":[177],"1.4":[179],"latency":[182],"improvement":[183],"compared":[184],"combination":[187],"typical":[189],"cutting-edge":[193],"Integer":[194],"ViT":[195],"module.":[196]},"counts_by_year":[],"updated_date":"2026-07-23T08:03:31.855105","created_date":"2026-04-10T00:00:00"}