{"id":"https://openalex.org/W4416922303","doi":"https://doi.org/10.1109/tcad.2025.3639392","title":"VersaAccel: A Versatile Configurable Accelerator for Diverse Sparse-Dense Matrix Operators","display_name":"VersaAccel: A Versatile Configurable Accelerator for Diverse Sparse-Dense Matrix Operators","publication_year":2025,"publication_date":"2025-12-02","ids":{"openalex":"https://openalex.org/W4416922303","doi":"https://doi.org/10.1109/tcad.2025.3639392"},"language":null,"primary_location":{"id":"doi:10.1109/tcad.2025.3639392","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tcad.2025.3639392","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/A5072361503","display_name":"Minjin Tang","orcid":"https://orcid.org/0009-0000-2254-4912"},"institutions":[{"id":"https://openalex.org/I2802541053","display_name":"Academy of Military Medical Sciences","ror":"https://ror.org/02bv3c993","country_code":"CN","type":"facility","lineage":["https://openalex.org/I2802541053"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Minjin Tang","raw_affiliation_strings":["Defense Innovation Institute, Academy of Military Science, Beijing, China","Defense Innovation Institute, Academy of Military Science, China"],"raw_orcid":"https://orcid.org/0009-0000-2254-4912","affiliations":[{"raw_affiliation_string":"Defense Innovation Institute, Academy of Military Science, Beijing, China","institution_ids":[]},{"raw_affiliation_string":"Defense Innovation Institute, Academy of Military Science, China","institution_ids":["https://openalex.org/I2802541053"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5101937502","display_name":"Mei Wen","orcid":"https://orcid.org/0000-0002-5875-3297"},"institutions":[{"id":"https://openalex.org/I111149068","display_name":"National Defense University","ror":"https://ror.org/01nqk4x38","country_code":"US","type":"education","lineage":["https://openalex.org/I111149068"]},{"id":"https://openalex.org/I170215575","display_name":"National University of Defense Technology","ror":"https://ror.org/05d2yfz11","country_code":"CN","type":"education","lineage":["https://openalex.org/I170215575"]}],"countries":["CN","US"],"is_corresponding":false,"raw_author_name":"Mei Wen","raw_affiliation_strings":["Key Laboratory of Advanced Microprocessor Chips and Systems, National University of Defense Technology, Changsha, China","Key Laboratory of Advanced Microprocessor Chips and Systems, National University of Defense Technology, China"],"raw_orcid":"https://orcid.org/0000-0002-5875-3297","affiliations":[{"raw_affiliation_string":"Key Laboratory of Advanced Microprocessor Chips and Systems, National University of Defense Technology, Changsha, China","institution_ids":["https://openalex.org/I170215575"]},{"raw_affiliation_string":"Key Laboratory of Advanced Microprocessor Chips and Systems, National University of Defense Technology, China","institution_ids":["https://openalex.org/I111149068"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":1.6796,"has_fulltext":false,"cited_by_count":1,"citation_normalized_percentile":{"value":0.89398404,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":95,"max":98},"biblio":{"volume":"45","issue":"7","first_page":"3059","last_page":"3072"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11612","display_name":"Stochastic Gradient Optimization Techniques","score":0.3163999915122986,"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"}},"topics":[{"id":"https://openalex.org/T11612","display_name":"Stochastic Gradient Optimization Techniques","score":0.3163999915122986,"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/T10036","display_name":"Advanced Neural Network Applications","score":0.20010000467300415,"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/T10363","display_name":"Low-power high-performance VLSI design","score":0.042899999767541885,"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/flexibility","display_name":"Flexibility (engineering)","score":0.6039999723434448},{"id":"https://openalex.org/keywords/set","display_name":"Set (abstract data type)","score":0.5497999787330627},{"id":"https://openalex.org/keywords/key","display_name":"Key (lock)","score":0.5437999963760376},{"id":"https://openalex.org/keywords/operator","display_name":"Operator (biology)","score":0.5169000029563904},{"id":"https://openalex.org/keywords/pruning","display_name":"Pruning","score":0.45590001344680786},{"id":"https://openalex.org/keywords/sparse-matrix","display_name":"Sparse matrix","score":0.45559999346733093},{"id":"https://openalex.org/keywords/energy","display_name":"Energy (signal processing)","score":0.4325999915599823}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7120000123977661},{"id":"https://openalex.org/C2780598303","wikidata":"https://www.wikidata.org/wiki/Q65921492","display_name":"Flexibility (engineering)","level":2,"score":0.6039999723434448},{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"score":0.5497999787330627},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.5437999963760376},{"id":"https://openalex.org/C17020691","wikidata":"https://www.wikidata.org/wiki/Q139677","display_name":"Operator (biology)","level":5,"score":0.5169000029563904},{"id":"https://openalex.org/C108010975","wikidata":"https://www.wikidata.org/wiki/Q500094","display_name":"Pruning","level":2,"score":0.45590001344680786},{"id":"https://openalex.org/C56372850","wikidata":"https://www.wikidata.org/wiki/Q1050404","display_name":"Sparse matrix","level":3,"score":0.45559999346733093},{"id":"https://openalex.org/C186370098","wikidata":"https://www.wikidata.org/wiki/Q442787","display_name":"Energy (signal processing)","level":2,"score":0.4325999915599823},{"id":"https://openalex.org/C2742236","wikidata":"https://www.wikidata.org/wiki/Q924713","display_name":"Efficient energy use","level":2,"score":0.4011000096797943},{"id":"https://openalex.org/C106487976","wikidata":"https://www.wikidata.org/wiki/Q685816","display_name":"Matrix (chemical analysis)","level":2,"score":0.39730000495910645},{"id":"https://openalex.org/C81917197","wikidata":"https://www.wikidata.org/wiki/Q628760","display_name":"Selection (genetic algorithm)","level":2,"score":0.3882000148296356},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.35440000891685486},{"id":"https://openalex.org/C17349429","wikidata":"https://www.wikidata.org/wiki/Q1049914","display_name":"Matrix multiplication","level":3,"score":0.3269999921321869},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.3231000006198883},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.3100999891757965},{"id":"https://openalex.org/C2780165032","wikidata":"https://www.wikidata.org/wiki/Q16869822","display_name":"Energy consumption","level":2,"score":0.29010000824928284},{"id":"https://openalex.org/C2779960059","wikidata":"https://www.wikidata.org/wiki/Q7113681","display_name":"Overhead (engineering)","level":2,"score":0.288100004196167},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.2624000012874603},{"id":"https://openalex.org/C42355184","wikidata":"https://www.wikidata.org/wiki/Q1361088","display_name":"Matrix decomposition","level":3,"score":0.2533999979496002}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/tcad.2025.3639392","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tcad.2025.3639392","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":[],"awards":[{"id":"https://openalex.org/G3421664612","display_name":null,"funder_award_id":"2024JJ6470","funder_id":"https://openalex.org/F4320322843","funder_display_name":"Natural Science Foundation of\u00a0Hunan Province"}],"funders":[{"id":"https://openalex.org/F4320322843","display_name":"Natural Science Foundation of\u00a0Hunan Province","ror":null}],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Matrix":[0],"operators":[1,41,52,156],"are":[2,37],"fundamental":[3],"to":[4,25,119,173],"various":[5],"applications,":[6],"particularly":[7],"in":[8,45],"deep":[9],"learning.":[10],"While":[11],"early":[12],"models":[13],"relied":[14],"on":[15,164],"dense":[16,29,74],"operations,":[17],"techniques":[18],"like":[19],"pruning":[20],"have":[21],"introduced":[22],"sparsity,":[23],"leading":[24],"a":[26,68,85,103],"mix":[27],"of":[28,87,92,152],"and":[30,42,56,73,162],"sparse":[31,72],"operator":[32],"types.":[33],"Most":[34],"existing":[35],"accelerators":[36],"specialized":[38],"for":[39,71,84,126],"specific":[40],"perform":[43],"poorly":[44],"mixed":[46],"scenarios,":[47],"while":[48],"those":[49],"supporting":[50],"multiple":[51,155],"often":[53],"lack":[54],"flexibility":[55],"suffer":[57],"from":[58],"suboptimal":[59],"performance.":[60],"To":[61],"overcome":[62],"these":[63],"limitations,":[64],"we":[65],"propose":[66],"VersaAccel,":[67],"configurable":[69],"accelerator":[70],"matrix":[75],"operators.":[76,88],"It":[77],"supports":[78],"four":[79],"distinct":[80],"configurations,":[81],"each":[82],"optimized":[83],"set":[86],"A":[89],"key":[90],"feature":[91],"our":[93],"design":[94],"is":[95],"its":[96],"adaptive":[97],"configuration":[98],"selection":[99],"mechanism,":[100],"driven":[101],"by":[102],"lightweight":[104],"cost":[105],"model":[106,166],"that":[107,145],"explicitly":[108],"evaluates":[109],"the":[110,122,130,133],"performance-energy":[111],"trade-off":[112],"between":[113],"available":[114],"options.":[115],"This":[116],"allows":[117],"VersaAccel":[118,146],"dynamically":[120],"choose":[121],"most":[123],"efficient":[124],"configuration\u2014opting":[125],"higher":[127],"performance":[128],"when":[129,140],"gain":[131],"outweighs":[132],"energy":[134,138],"cost,":[135],"or":[136],"prioritizing":[137],"efficiency":[139],"appropriate.":[141],"Experimental":[142],"results":[143],"demonstrate":[144],"achieves":[147],"an":[148],"average":[149],"performance/area":[150],"improvement":[151],"3.10\u00d7":[153],"across":[154],"(MV,":[157],"MM,":[158],"SPMV,":[159],"SpMM,":[160],"SpMSpM)":[161],"4.06\u00d7":[163],"full":[165],"evaluations":[167],"(ResNet18,":[168],"VGG16,":[169],"LLaMA2-7B,":[170],"BERT-Base),":[171],"compared":[172],"mainstream":[174],"accelerators.":[175]},"counts_by_year":[{"year":2026,"cited_by_count":1}],"updated_date":"2026-06-20T20:08:15.867695","created_date":"2025-12-02T00:00:00"}