{"id":"https://openalex.org/W4411486557","doi":"https://doi.org/10.1145/3695053.3731043","title":"Transitive Array: An Efficient GEMM Accelerator with Result Reuse","display_name":"Transitive Array: An Efficient GEMM Accelerator with Result Reuse","publication_year":2025,"publication_date":"2025-06-20","ids":{"openalex":"https://openalex.org/W4411486557","doi":"https://doi.org/10.1145/3695053.3731043"},"language":"en","primary_location":{"id":"doi:10.1145/3695053.3731043","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3695053.3731043","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3695053.3731043","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 52nd Annual International Symposium on Computer Architecture","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://dl.acm.org/doi/pdf/10.1145/3695053.3731043","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5101706991","display_name":"Cong Guo","orcid":"https://orcid.org/0000-0002-4479-5525"},"institutions":[{"id":"https://openalex.org/I170897317","display_name":"Duke University","ror":"https://ror.org/00py81415","country_code":"US","type":"education","lineage":["https://openalex.org/I170897317"]}],"countries":["US"],"is_corresponding":true,"raw_author_name":"Cong Guo","raw_affiliation_strings":["Duke University, Durham, USA"],"affiliations":[{"raw_affiliation_string":"Duke University, Durham, USA","institution_ids":["https://openalex.org/I170897317"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5060545776","display_name":"Chiyue Wei","orcid":"https://orcid.org/0009-0008-8815-7948"},"institutions":[{"id":"https://openalex.org/I170897317","display_name":"Duke University","ror":"https://ror.org/00py81415","country_code":"US","type":"education","lineage":["https://openalex.org/I170897317"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Chiyue Wei","raw_affiliation_strings":["Duke University, Durham, USA"],"affiliations":[{"raw_affiliation_string":"Duke University, Durham, USA","institution_ids":["https://openalex.org/I170897317"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5103244882","display_name":"Jiaming Tang","orcid":"https://orcid.org/0009-0004-4186-6561"},"institutions":[{"id":"https://openalex.org/I63966007","display_name":"Massachusetts Institute of Technology","ror":"https://ror.org/042nb2s44","country_code":"US","type":"education","lineage":["https://openalex.org/I63966007"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Jiaming Tang","raw_affiliation_strings":["Massachusetts Institute of Technology, Cambridge, USA"],"affiliations":[{"raw_affiliation_string":"Massachusetts Institute of Technology, Cambridge, USA","institution_ids":["https://openalex.org/I63966007"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5116907275","display_name":"Bowen Duan","orcid":"https://orcid.org/0009-0004-9085-5025"},"institutions":[{"id":"https://openalex.org/I170897317","display_name":"Duke University","ror":"https://ror.org/00py81415","country_code":"US","type":"education","lineage":["https://openalex.org/I170897317"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Bowen Duan","raw_affiliation_strings":["Duke University, Durham, USA"],"affiliations":[{"raw_affiliation_string":"Duke University, Durham, USA","institution_ids":["https://openalex.org/I170897317"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5070926896","display_name":"Song Han","orcid":"https://orcid.org/0000-0002-4186-7618"},"institutions":[{"id":"https://openalex.org/I63966007","display_name":"Massachusetts Institute of Technology","ror":"https://ror.org/042nb2s44","country_code":"US","type":"education","lineage":["https://openalex.org/I63966007"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Song Han","raw_affiliation_strings":["Massachusetts Institute of Technology, Cambridge, USA"],"affiliations":[{"raw_affiliation_string":"Massachusetts Institute of Technology, Cambridge, USA","institution_ids":["https://openalex.org/I63966007"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100429403","display_name":"Hai Li","orcid":"https://orcid.org/0000-0003-3228-6544"},"institutions":[{"id":"https://openalex.org/I170897317","display_name":"Duke University","ror":"https://ror.org/00py81415","country_code":"US","type":"education","lineage":["https://openalex.org/I170897317"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Hai Li","raw_affiliation_strings":["Duke University, Durham, USA"],"affiliations":[{"raw_affiliation_string":"Duke University, Durham, USA","institution_ids":["https://openalex.org/I170897317"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5058073627","display_name":"Yiran Chen","orcid":"https://orcid.org/0000-0002-1486-8412"},"institutions":[{"id":"https://openalex.org/I170897317","display_name":"Duke University","ror":"https://ror.org/00py81415","country_code":"US","type":"education","lineage":["https://openalex.org/I170897317"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Yiran Chen","raw_affiliation_strings":["Duke University, Durham, USA"],"affiliations":[{"raw_affiliation_string":"Duke University, Durham, USA","institution_ids":["https://openalex.org/I170897317"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":7,"corresponding_author_ids":["https://openalex.org/A5101706991"],"corresponding_institution_ids":["https://openalex.org/I170897317"],"apc_list":null,"apc_paid":null,"fwci":12.7854,"has_fulltext":true,"cited_by_count":7,"citation_normalized_percentile":{"value":0.98406889,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":95,"max":100},"biblio":{"volume":null,"issue":null,"first_page":"990","last_page":"1004"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11044","display_name":"Particle Detector Development and Performance","score":0.9829999804496765,"subfield":{"id":"https://openalex.org/subfields/3106","display_name":"Nuclear and High Energy Physics"},"field":{"id":"https://openalex.org/fields/31","display_name":"Physics and Astronomy"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T11044","display_name":"Particle Detector Development and Performance","score":0.9829999804496765,"subfield":{"id":"https://openalex.org/subfields/3106","display_name":"Nuclear and High Energy Physics"},"field":{"id":"https://openalex.org/fields/31","display_name":"Physics and Astronomy"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T11367","display_name":"Particle accelerators and beam dynamics","score":0.9786999821662903,"subfield":{"id":"https://openalex.org/subfields/2202","display_name":"Aerospace 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/T11392","display_name":"Energy Harvesting in Wireless Networks","score":0.9728000164031982,"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/computer-science","display_name":"Computer science","score":0.6797240972518921},{"id":"https://openalex.org/keywords/transitive-relation","display_name":"Transitive relation","score":0.5695580840110779},{"id":"https://openalex.org/keywords/reuse","display_name":"Reuse","score":0.5377490520477295},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.5363795757293701},{"id":"https://openalex.org/keywords/programming-language","display_name":"Programming language","score":0.4025001525878906},{"id":"https://openalex.org/keywords/computational-science","display_name":"Computational science","score":0.3570583462715149},{"id":"https://openalex.org/keywords/computer-architecture","display_name":"Computer architecture","score":0.34268778562545776},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.22010335326194763},{"id":"https://openalex.org/keywords/engineering","display_name":"Engineering","score":0.09646570682525635},{"id":"https://openalex.org/keywords/combinatorics","display_name":"Combinatorics","score":0.07798391580581665}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6797240972518921},{"id":"https://openalex.org/C191399111","wikidata":"https://www.wikidata.org/wiki/Q64861","display_name":"Transitive relation","level":2,"score":0.5695580840110779},{"id":"https://openalex.org/C206588197","wikidata":"https://www.wikidata.org/wiki/Q846574","display_name":"Reuse","level":2,"score":0.5377490520477295},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.5363795757293701},{"id":"https://openalex.org/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.4025001525878906},{"id":"https://openalex.org/C459310","wikidata":"https://www.wikidata.org/wiki/Q117801","display_name":"Computational science","level":1,"score":0.3570583462715149},{"id":"https://openalex.org/C118524514","wikidata":"https://www.wikidata.org/wiki/Q173212","display_name":"Computer architecture","level":1,"score":0.34268778562545776},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.22010335326194763},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.09646570682525635},{"id":"https://openalex.org/C114614502","wikidata":"https://www.wikidata.org/wiki/Q76592","display_name":"Combinatorics","level":1,"score":0.07798391580581665},{"id":"https://openalex.org/C548081761","wikidata":"https://www.wikidata.org/wiki/Q180388","display_name":"Waste management","level":1,"score":0.0}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1145/3695053.3731043","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3695053.3731043","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3695053.3731043","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 52nd Annual International Symposium on Computer Architecture","raw_type":"proceedings-article"},{"id":"pmh:oai:dspace.mit.edu:1721.1/162666","is_oa":true,"landing_page_url":"https://hdl.handle.net/1721.1/162666","pdf_url":"https://dspace.mit.edu/bitstream/1721.1/162666/1/3695053.3731043.pdf","source":{"id":"https://openalex.org/S4306400425","display_name":"DSpace@MIT (Massachusetts Institute of Technology)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I63966007","host_organization_name":"Massachusetts Institute of Technology","host_organization_lineage":["https://openalex.org/I63966007"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by-nc","license_id":"https://openalex.org/licenses/cc-by-nc","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"Association for Computing Machinery","raw_type":"http://purl.org/eprint/type/ConferencePaper"}],"best_oa_location":{"id":"doi:10.1145/3695053.3731043","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3695053.3731043","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3695053.3731043","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 52nd Annual International Symposium on Computer Architecture","raw_type":"proceedings-article"},"sustainable_development_goals":[{"score":0.41999998688697815,"display_name":"Affordable and clean energy","id":"https://metadata.un.org/sdg/7"}],"awards":[{"id":"https://openalex.org/G6624552157","display_name":null,"funder_award_id":"W911NF-23-2-0224","funder_id":"https://openalex.org/F4320338281","funder_display_name":"Army Research Office"},{"id":"https://openalex.org/G7452299184","display_name":null,"funder_award_id":"W911NF","funder_id":"https://openalex.org/F4320338281","funder_display_name":"Army Research Office"},{"id":"https://openalex.org/G8377316191","display_name":null,"funder_award_id":"2112562","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"},{"id":"https://openalex.org/G848032724","display_name":null,"funder_award_id":"Science","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"},{"id":"https://openalex.org/G8998121839","display_name":null,"funder_award_id":"911NF","funder_id":"https://openalex.org/F4320338281","funder_display_name":"Army Research Office"}],"funders":[{"id":"https://openalex.org/F4320306076","display_name":"National Science Foundation","ror":"https://ror.org/021nxhr62"},{"id":"https://openalex.org/F4320338281","display_name":"Army Research Office","ror":"https://ror.org/05epdh915"}],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4411486557.pdf","grobid_xml":"https://content.openalex.org/works/W4411486557.grobid-xml"},"referenced_works_count":31,"referenced_works":["https://openalex.org/W2011849452","https://openalex.org/W2094969361","https://openalex.org/W2108598243","https://openalex.org/W2141389982","https://openalex.org/W2194775991","https://openalex.org/W2285660444","https://openalex.org/W2508602506","https://openalex.org/W2513721464","https://openalex.org/W2541839172","https://openalex.org/W2611106620","https://openalex.org/W2752908210","https://openalex.org/W2883920103","https://openalex.org/W2919115771","https://openalex.org/W2953212265","https://openalex.org/W2963367920","https://openalex.org/W2980186997","https://openalex.org/W2990138404","https://openalex.org/W3016542674","https://openalex.org/W3100985894","https://openalex.org/W3187908937","https://openalex.org/W4308083739","https://openalex.org/W4366341968","https://openalex.org/W4391827186","https://openalex.org/W4393406875","https://openalex.org/W4401211807","https://openalex.org/W4403211793","https://openalex.org/W4404954664","https://openalex.org/W4407212618","https://openalex.org/W4409248468","https://openalex.org/W4409248487","https://openalex.org/W4409248600"],"related_works":["https://openalex.org/W4312527695","https://openalex.org/W2361167282","https://openalex.org/W1528932152","https://openalex.org/W2091342995","https://openalex.org/W2271118953","https://openalex.org/W3007067598","https://openalex.org/W2359420171","https://openalex.org/W1677394555","https://openalex.org/W2073498251","https://openalex.org/W2006385248"],"abstract_inverted_index":{"Deep":[0],"Neural":[1],"Networks":[2],"(DNNs)":[3],"and":[4,19,94,126,139,142,144,155],"Large":[5],"Language":[6],"Models":[7],"(LLMs)":[8],"have":[9,27],"revolutionized":[10],"artificial":[11],"intelligence,":[12],"yet":[13],"their":[14],"deployment":[15],"faces":[16],"significant":[17],"memory":[18],"computational":[20,63,117],"challenges,":[21],"especially":[22],"in":[23,65,112],"resource-constrained":[24],"environments.Quantization":[25],"techniques":[26],"mitigated":[28],"some":[29],"of":[30,56],"these":[31],"issues":[32],"by":[33],"reducing":[34],"data":[35],"precision,":[36],"primarily":[37],"focusing":[38],"on":[39,96,162],"General":[40],"Matrix":[41],"Multiplication":[42],"(GEMM).This":[43],"study":[44],"introduces":[45],"a":[46,72,104],"novel":[47],"sparsity":[48,111],"paradigm,":[49],"transitive":[50,69,110],"sparsity,":[51],"which":[52],"leverages":[53],"the":[54,101,133],"reuse":[55],"previously":[57],"computed":[58],"results":[59],"to":[60,91,108,149],"substantially":[61],"minimize":[62],"overhead":[64],"GEMM":[66],"operations.By":[67],"representing":[68],"relations":[70],"using":[71],"directed":[73],"acyclic":[74],"graph,":[75],"we":[76,99],"develop":[77],"an":[78],"efficient":[79],"strategy":[80],"for":[81],"determining":[82],"optimal":[83,127],"execution":[84,92],"orders,":[85],"thereby":[86],"overcoming":[87],"inherent":[88],"challenges":[89],"related":[90],"dependencies":[93],"parallelism.Building":[95],"this":[97],"foundation,":[98],"present":[100],"Transitive":[102,134],"Array,":[103],"multiplication-free":[105],"accelerator":[106],"designed":[107],"exploit":[109],"GEMM.Our":[113],"architecture":[114],"effectively":[115],"balances":[116],"workloads":[118],"across":[119],"multiple":[120],"parallel":[121],"lanes,":[122],"ensuring":[123],"high":[124],"efficiency":[125],"resource":[128],"utilization.Comprehensive":[129],"evaluations":[130],"demonstrate":[131],"that":[132],"Array":[135],"achieves":[136],"approximately":[137],"7.46":[138],"3.97":[140],"speedup":[141],"2.31":[143],"1.65":[145],"energy":[146],"reduction":[147],"compared":[148],"state-of-the-art":[150],"accelerators":[151],"such":[152],"as":[153],"Olive":[154],"BitVert":[156],"while":[157],"maintaining":[158],"comparable":[159],"model":[160],"accuracy":[161],"LLaMA":[163],"models.":[164]},"counts_by_year":[{"year":2026,"cited_by_count":5},{"year":2025,"cited_by_count":2}],"updated_date":"2026-04-04T16:13:02.066488","created_date":"2025-10-10T00:00:00"}