{"id":"https://openalex.org/W4417260734","doi":"https://doi.org/10.1145/3779444","title":"REATA: An Efficient Vision Transformer Accelerator Featuring a Resource-Optimized Attention Design on\u00a0Versal\u00a0ACAP","display_name":"REATA: An Efficient Vision Transformer Accelerator Featuring a Resource-Optimized Attention Design on\u00a0Versal\u00a0ACAP","publication_year":2025,"publication_date":"2025-12-12","ids":{"openalex":"https://openalex.org/W4417260734","doi":"https://doi.org/10.1145/3779444"},"language":"en","primary_location":{"id":"doi:10.1145/3779444","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3779444","pdf_url":null,"source":{"id":"https://openalex.org/S112809824","display_name":"ACM Transactions on Reconfigurable Technology and Systems","issn_l":"1936-7406","issn":["1936-7406","1936-7414"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319798","host_organization_name":"Association for Computing Machinery","host_organization_lineage":["https://openalex.org/P4310319798"],"host_organization_lineage_names":["Association for Computing Machinery"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ACM Transactions on Reconfigurable Technology 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/A5100398462","display_name":"Wenbo Zhang","orcid":"https://orcid.org/0000-0002-8601-802X"},"institutions":[{"id":"https://openalex.org/I37796252","display_name":"Beijing University of Technology","ror":"https://ror.org/037b1pp87","country_code":"CN","type":"education","lineage":["https://openalex.org/I37796252"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Wenbo Zhang","raw_affiliation_strings":["College of Computer Science, Beijing University of Technology, Beijing, China","Beijing University of Technology, China"],"raw_orcid":"https://orcid.org/0000-0002-8601-802X","affiliations":[{"raw_affiliation_string":"College of Computer Science, Beijing University of Technology, Beijing, China","institution_ids":["https://openalex.org/I37796252"]},{"raw_affiliation_string":"Beijing University of Technology, China","institution_ids":["https://openalex.org/I37796252"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Yan Zhang","orcid":"https://orcid.org/0009-0006-5628-6943"},"institutions":[{"id":"https://openalex.org/I37796252","display_name":"Beijing University of Technology","ror":"https://ror.org/037b1pp87","country_code":"CN","type":"education","lineage":["https://openalex.org/I37796252"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yan Zhang","raw_affiliation_strings":["College of Computer Science, Beijing University of Technology, Beijing, China","Beijing University of Technology, China"],"raw_orcid":"https://orcid.org/0009-0006-5628-6943","affiliations":[{"raw_affiliation_string":"College of Computer Science, Beijing University of Technology, Beijing, China","institution_ids":["https://openalex.org/I37796252"]},{"raw_affiliation_string":"Beijing University of Technology, China","institution_ids":["https://openalex.org/I37796252"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5087352252","display_name":"Yiqi Liu","orcid":"https://orcid.org/0009-0000-8262-2185"},"institutions":[{"id":"https://openalex.org/I37796252","display_name":"Beijing University of Technology","ror":"https://ror.org/037b1pp87","country_code":"CN","type":"education","lineage":["https://openalex.org/I37796252"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yiqi Liu","raw_affiliation_strings":["College of Computer Science, Beijing University of Technology, Beijing, China","Beijing University of Technology, China"],"raw_orcid":"https://orcid.org/0009-0000-8262-2185","affiliations":[{"raw_affiliation_string":"College of Computer Science, Beijing University of Technology, Beijing, China","institution_ids":["https://openalex.org/I37796252"]},{"raw_affiliation_string":"Beijing University of Technology, China","institution_ids":["https://openalex.org/I37796252"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Lingjie Wu","orcid":"https://orcid.org/0009-0000-2197-0928"},"institutions":[{"id":"https://openalex.org/I37796252","display_name":"Beijing University of Technology","ror":"https://ror.org/037b1pp87","country_code":"CN","type":"education","lineage":["https://openalex.org/I37796252"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Lingjie Wu","raw_affiliation_strings":["College of Computer Science, Beijing University of Technology, Beijing, China","Beijing University of Technology, China"],"raw_orcid":"https://orcid.org/0009-0000-2197-0928","affiliations":[{"raw_affiliation_string":"College of Computer Science, Beijing University of Technology, Beijing, China","institution_ids":["https://openalex.org/I37796252"]},{"raw_affiliation_string":"Beijing University of Technology, China","institution_ids":["https://openalex.org/I37796252"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5104342881","display_name":"Xiaokun Hu","orcid":"https://orcid.org/0009-0005-6281-8515"},"institutions":[{"id":"https://openalex.org/I37796252","display_name":"Beijing University of Technology","ror":"https://ror.org/037b1pp87","country_code":"CN","type":"education","lineage":["https://openalex.org/I37796252"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xingtong Hu","raw_affiliation_strings":["College of Computer Science, Beijing University of Technology, Beijing, China","Beijing University of Technology, China"],"raw_orcid":"https://orcid.org/0009-0005-6281-8515","affiliations":[{"raw_affiliation_string":"College of Computer Science, Beijing University of Technology, Beijing, China","institution_ids":["https://openalex.org/I37796252"]},{"raw_affiliation_string":"Beijing University of Technology, China","institution_ids":["https://openalex.org/I37796252"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I37796252"],"apc_list":null,"apc_paid":null,"fwci":1.5311,"has_fulltext":false,"cited_by_count":2,"citation_normalized_percentile":{"value":0.86084006,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":97,"max":98},"biblio":{"volume":"19","issue":"1","first_page":"1","last_page":"32"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.27970001101493835,"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":0.27970001101493835,"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/T10502","display_name":"Advanced Memory and Neural Computing","score":0.24289999902248383,"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/T11992","display_name":"CCD and CMOS Imaging Sensors","score":0.17560000717639923,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/scalability","display_name":"Scalability","score":0.6811000108718872},{"id":"https://openalex.org/keywords/modular-design","display_name":"Modular design","score":0.6488000154495239},{"id":"https://openalex.org/keywords/dataflow","display_name":"Dataflow","score":0.6421999931335449},{"id":"https://openalex.org/keywords/transformer","display_name":"Transformer","score":0.5187000036239624},{"id":"https://openalex.org/keywords/computation","display_name":"Computation","score":0.4832000136375427},{"id":"https://openalex.org/keywords/hardware-acceleration","display_name":"Hardware acceleration","score":0.4805000126361847},{"id":"https://openalex.org/keywords/architecture","display_name":"Architecture","score":0.4122999906539917},{"id":"https://openalex.org/keywords/edge-device","display_name":"Edge device","score":0.40119999647140503},{"id":"https://openalex.org/keywords/field-programmable-gate-array","display_name":"Field-programmable gate array","score":0.3991999924182892}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.866100013256073},{"id":"https://openalex.org/C48044578","wikidata":"https://www.wikidata.org/wiki/Q727490","display_name":"Scalability","level":2,"score":0.6811000108718872},{"id":"https://openalex.org/C118524514","wikidata":"https://www.wikidata.org/wiki/Q173212","display_name":"Computer architecture","level":1,"score":0.6653000116348267},{"id":"https://openalex.org/C101468663","wikidata":"https://www.wikidata.org/wiki/Q1620158","display_name":"Modular design","level":2,"score":0.6488000154495239},{"id":"https://openalex.org/C96324660","wikidata":"https://www.wikidata.org/wiki/Q205446","display_name":"Dataflow","level":2,"score":0.6421999931335449},{"id":"https://openalex.org/C149635348","wikidata":"https://www.wikidata.org/wiki/Q193040","display_name":"Embedded system","level":1,"score":0.5841000080108643},{"id":"https://openalex.org/C66322947","wikidata":"https://www.wikidata.org/wiki/Q11658","display_name":"Transformer","level":3,"score":0.5187000036239624},{"id":"https://openalex.org/C45374587","wikidata":"https://www.wikidata.org/wiki/Q12525525","display_name":"Computation","level":2,"score":0.4832000136375427},{"id":"https://openalex.org/C13164978","wikidata":"https://www.wikidata.org/wiki/Q600158","display_name":"Hardware acceleration","level":3,"score":0.4805000126361847},{"id":"https://openalex.org/C123657996","wikidata":"https://www.wikidata.org/wiki/Q12271","display_name":"Architecture","level":2,"score":0.4122999906539917},{"id":"https://openalex.org/C138236772","wikidata":"https://www.wikidata.org/wiki/Q25098575","display_name":"Edge device","level":3,"score":0.40119999647140503},{"id":"https://openalex.org/C42935608","wikidata":"https://www.wikidata.org/wiki/Q190411","display_name":"Field-programmable gate array","level":2,"score":0.3991999924182892},{"id":"https://openalex.org/C43521106","wikidata":"https://www.wikidata.org/wiki/Q2165493","display_name":"Pipeline (software)","level":2,"score":0.38089999556541443},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.3797999918460846},{"id":"https://openalex.org/C206729178","wikidata":"https://www.wikidata.org/wiki/Q2271896","display_name":"Scheduling (production processes)","level":2,"score":0.373199999332428},{"id":"https://openalex.org/C2778456923","wikidata":"https://www.wikidata.org/wiki/Q5337692","display_name":"Edge computing","level":3,"score":0.3578000068664551},{"id":"https://openalex.org/C177606310","wikidata":"https://www.wikidata.org/wiki/Q5674297","display_name":"Adaptability","level":2,"score":0.35760000348091125},{"id":"https://openalex.org/C9390403","wikidata":"https://www.wikidata.org/wiki/Q3966","display_name":"Computer hardware","level":1,"score":0.34929999709129333},{"id":"https://openalex.org/C79974875","wikidata":"https://www.wikidata.org/wiki/Q483639","display_name":"Cloud computing","level":2,"score":0.3197999894618988},{"id":"https://openalex.org/C142962650","wikidata":"https://www.wikidata.org/wiki/Q240838","display_name":"Reconfigurable computing","level":3,"score":0.3021000027656555},{"id":"https://openalex.org/C206588197","wikidata":"https://www.wikidata.org/wiki/Q846574","display_name":"Reuse","level":2,"score":0.29280000925064087},{"id":"https://openalex.org/C107027933","wikidata":"https://www.wikidata.org/wiki/Q2006448","display_name":"Stream processing","level":2,"score":0.2888000011444092},{"id":"https://openalex.org/C2742236","wikidata":"https://www.wikidata.org/wiki/Q924713","display_name":"Efficient energy use","level":2,"score":0.28870001435279846},{"id":"https://openalex.org/C2777210771","wikidata":"https://www.wikidata.org/wiki/Q4927124","display_name":"Block (permutation group theory)","level":2,"score":0.2736999988555908},{"id":"https://openalex.org/C157764524","wikidata":"https://www.wikidata.org/wiki/Q1383412","display_name":"Throughput","level":3,"score":0.2694999873638153},{"id":"https://openalex.org/C200106649","wikidata":"https://www.wikidata.org/wiki/Q223683","display_name":"Transpose","level":3,"score":0.2685000002384186},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.26669999957084656}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3779444","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3779444","pdf_url":null,"source":{"id":"https://openalex.org/S112809824","display_name":"ACM Transactions on Reconfigurable Technology and Systems","issn_l":"1936-7406","issn":["1936-7406","1936-7414"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319798","host_organization_name":"Association for Computing Machinery","host_organization_lineage":["https://openalex.org/P4310319798"],"host_organization_lineage_names":["Association for Computing Machinery"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ACM Transactions on Reconfigurable Technology and Systems","raw_type":"journal-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":25,"referenced_works":["https://openalex.org/W2606722458","https://openalex.org/W2918037051","https://openalex.org/W2963351113","https://openalex.org/W2981712449","https://openalex.org/W3096609285","https://openalex.org/W3138516171","https://openalex.org/W3211328899","https://openalex.org/W4214612132","https://openalex.org/W4225672218","https://openalex.org/W4308479898","https://openalex.org/W4319870545","https://openalex.org/W4321637298","https://openalex.org/W4360831786","https://openalex.org/W4360831795","https://openalex.org/W4390873361","https://openalex.org/W4391125526","https://openalex.org/W4393949386","https://openalex.org/W4393973515","https://openalex.org/W4400656480","https://openalex.org/W4401331008","https://openalex.org/W4404102506","https://openalex.org/W4409261789","https://openalex.org/W4409262823","https://openalex.org/W4409307140","https://openalex.org/W4413145332"],"related_works":[],"abstract_inverted_index":{"Deploying":[0],"Vision":[1],"Transformers":[2],"(ViTs)":[3],"on":[4,129,162],"edge":[5,171],"devices":[6],"poses":[7],"significant":[8],"challenges":[9],"due":[10],"to":[11,117],"their":[12],"high":[13],"computational":[14],"demands":[15],"and":[16,31,47,79,94,104,112,164,176],"memory":[17,125],"access":[18],"overheads,":[19],"which":[20],"severely":[21],"hinder":[22],"real-time":[23],"inference":[24],"efficiency.":[25],"This":[26],"article":[27],"proposes":[28],"a":[29,60,86,113,119,154,174],"modular":[30],"adaptive":[32],"ViT":[33],"acceleration":[34],"architecture":[35,108],"targeting":[36],"the":[37,51,96,130,137,147,167],"AMD":[38],"Versal":[39],"ACAP":[40,133],"platform.":[41],"By":[42],"leveraging":[43],"heterogeneous":[44],"resource":[45,82],"collaboration":[46],"fine-grained":[48],"dataflow":[49,122],"optimizations,":[50],"proposed":[52,138],"design":[53,139],"addresses":[54],"performance":[55],"bottlenecks":[56],"effectively.":[57],"We":[58],"introduce":[59],"resource-efficient":[61,175],"attention":[62],"computation":[63,102],"module":[64],"that":[65,136],"localizes":[66],"self-attention":[67],"operations":[68],"within":[69],"AI":[70],"Engine":[71],"(AIE)":[72],"core":[73],"clusters,":[74],"thereby":[75],"reducing":[76],"inter-module":[77],"communication":[78],"minimizing":[80],"MAC":[81],"usage.":[83],"In":[84],"parallel,":[85],"resource-aware":[87],"multi-stage":[88],"pipeline":[89],"scheduling":[90],"strategy":[91],"dynamically":[92],"partitions":[93],"parallelizes":[95],"computation-intensive":[97],"feed-forward":[98],"network":[99],"(FFN),":[100],"improving":[101],"reuse":[103],"module-level":[105],"coordination.":[106],"The":[107],"integrates":[109],"parameter":[110],"tiling":[111],"PLIO-based":[114],"broadcasting":[115],"mechanism":[116],"construct":[118],"decoupled":[120],"compute-communication":[121],"engine,":[123],"alleviating":[124],"bottlenecks.":[126],"Experimental":[127],"results":[128],"Xilinx":[131],"VCK5000":[132],"platform":[134],"demonstrate":[135],"achieves":[140],"33.2":[141],"TOPS":[142],"throughput":[143],"at":[144],"INT8":[145],"precision\u2014outperforming":[146],"state-of-the-art":[148],"EQ-ViT":[149],"accelerator":[150],"by":[151],"27%\u2014while":[152],"maintaining":[153],"competitive":[155],"efficiency":[156],"of":[157],"510.6":[158],"GOPS/W.":[159],"Scalability":[160],"evaluations":[161],"ViT-Base":[163],"DeiT-Tiny":[165],"confirm":[166],"design\u2019s":[168],"adaptability":[169],"in":[170],"scenarios,":[172],"offering":[173],"reconfigurable":[177],"hardware":[178],"paradigm":[179],"for":[180],"high-density":[181],"Transformer":[182],"inference.":[183]},"counts_by_year":[{"year":2026,"cited_by_count":2}],"updated_date":"2026-05-21T06:26:12.895304","created_date":"2025-12-12T00:00:00"}