{"id":"https://openalex.org/W7165157857","doi":"https://doi.org/10.1145/3787109.3815235","title":"BitPair: An Efficient 2-Bit Serial Precision-Scalable Accelerator for GEMM in Deep Neural Networks","display_name":"BitPair: An Efficient 2-Bit Serial Precision-Scalable Accelerator for GEMM in Deep Neural Networks","publication_year":2026,"publication_date":"2026-06-18","ids":{"openalex":"https://openalex.org/W7165157857","doi":"https://doi.org/10.1145/3787109.3815235"},"language":null,"primary_location":{"id":"doi:10.1145/3787109.3815235","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3787109.3815235","pdf_url":null,"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 Great Lakes Symposium on VLSI 2026","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://doi.org/10.1145/3787109.3815235","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5100319327","display_name":"Jinhua Li","orcid":"https://orcid.org/0000-0002-5226-0272"},"institutions":[{"id":"https://openalex.org/I24943067","display_name":"Fudan University","ror":"https://ror.org/013q1eq08","country_code":"CN","type":"education","lineage":["https://openalex.org/I24943067"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Jinhua Li","raw_affiliation_strings":["State Key Laboratory of Integrated Chips and Systems, Fudan University, Shanghai, Shanghai, China"],"raw_orcid":"https://orcid.org/0009-0009-9498-0428","affiliations":[{"raw_affiliation_string":"State Key Laboratory of Integrated Chips and Systems, Fudan University, Shanghai, Shanghai, China","institution_ids":["https://openalex.org/I24943067"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100743030","display_name":"Menghan Li","orcid":"https://orcid.org/0009-0001-9796-6471"},"institutions":[{"id":"https://openalex.org/I24943067","display_name":"Fudan University","ror":"https://ror.org/013q1eq08","country_code":"CN","type":"education","lineage":["https://openalex.org/I24943067"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Menghan Li","raw_affiliation_strings":["State Key Laboratory of Integrated Chips and Systems, Fudan University, Shanghai, Shanghai, China"],"raw_orcid":"https://orcid.org/0009-0001-9796-6471","affiliations":[{"raw_affiliation_string":"State Key Laboratory of Integrated Chips and Systems, Fudan University, Shanghai, Shanghai, China","institution_ids":["https://openalex.org/I24943067"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5062927280","display_name":"Jun Tao","orcid":"https://orcid.org/0000-0001-8742-687X"},"institutions":[{"id":"https://openalex.org/I24943067","display_name":"Fudan University","ror":"https://ror.org/013q1eq08","country_code":"CN","type":"education","lineage":["https://openalex.org/I24943067"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Jun Tao","raw_affiliation_strings":["State Key Laboratory of Integrated Chips and Systems, Fudan University, Shanghai, Shanghai, China"],"raw_orcid":"https://orcid.org/0000-0001-8742-687X","affiliations":[{"raw_affiliation_string":"State Key Laboratory of Integrated Chips and Systems, Fudan University, Shanghai, Shanghai, China","institution_ids":["https://openalex.org/I24943067"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5066674438","display_name":"Jun Han","orcid":"https://orcid.org/0000-0002-5245-0754"},"institutions":[{"id":"https://openalex.org/I24943067","display_name":"Fudan University","ror":"https://ror.org/013q1eq08","country_code":"CN","type":"education","lineage":["https://openalex.org/I24943067"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Jun Han","raw_affiliation_strings":["State Key Laboratory of Integrated Chips and Systems, Fudan University, Shanghai, Shanghai, China"],"raw_orcid":"https://orcid.org/0000-0002-5245-0754","affiliations":[{"raw_affiliation_string":"State Key Laboratory of Integrated Chips and Systems, Fudan University, Shanghai, Shanghai, China","institution_ids":["https://openalex.org/I24943067"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I24943067"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.75291025,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"392","last_page":"397"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11697","display_name":"Numerical Methods and Algorithms","score":0.474700003862381,"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"}},"topics":[{"id":"https://openalex.org/T11697","display_name":"Numerical Methods and Algorithms","score":0.474700003862381,"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"}},{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.1835000067949295,"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/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.03830000013113022,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/dataflow","display_name":"Dataflow","score":0.7310000061988831},{"id":"https://openalex.org/keywords/quantization","display_name":"Quantization (signal processing)","score":0.6870999932289124},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.6158000230789185},{"id":"https://openalex.org/keywords/computation","display_name":"Computation","score":0.5860999822616577},{"id":"https://openalex.org/keywords/deep-neural-networks","display_name":"Deep neural networks","score":0.5175999999046326},{"id":"https://openalex.org/keywords/integer","display_name":"Integer (computer science)","score":0.5105000138282776},{"id":"https://openalex.org/keywords/verilog","display_name":"Verilog","score":0.41269999742507935}],"concepts":[{"id":"https://openalex.org/C96324660","wikidata":"https://www.wikidata.org/wiki/Q205446","display_name":"Dataflow","level":2,"score":0.7310000061988831},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7128000259399414},{"id":"https://openalex.org/C28855332","wikidata":"https://www.wikidata.org/wiki/Q198099","display_name":"Quantization (signal processing)","level":2,"score":0.6870999932289124},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.6158000230789185},{"id":"https://openalex.org/C45374587","wikidata":"https://www.wikidata.org/wiki/Q12525525","display_name":"Computation","level":2,"score":0.5860999822616577},{"id":"https://openalex.org/C2984842247","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep neural networks","level":3,"score":0.5175999999046326},{"id":"https://openalex.org/C97137487","wikidata":"https://www.wikidata.org/wiki/Q729138","display_name":"Integer (computer science)","level":2,"score":0.5105000138282776},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.49970000982284546},{"id":"https://openalex.org/C2779030575","wikidata":"https://www.wikidata.org/wiki/Q827773","display_name":"Verilog","level":3,"score":0.41269999742507935},{"id":"https://openalex.org/C2778584943","wikidata":"https://www.wikidata.org/wiki/Q6459541","display_name":"LOOM","level":2,"score":0.39430001378059387},{"id":"https://openalex.org/C106487976","wikidata":"https://www.wikidata.org/wiki/Q685816","display_name":"Matrix (chemical analysis)","level":2,"score":0.36739999055862427},{"id":"https://openalex.org/C17349429","wikidata":"https://www.wikidata.org/wiki/Q1049914","display_name":"Matrix multiplication","level":3,"score":0.3610999882221222},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.3409000039100647},{"id":"https://openalex.org/C186370098","wikidata":"https://www.wikidata.org/wiki/Q442787","display_name":"Energy (signal processing)","level":2,"score":0.33160001039505005},{"id":"https://openalex.org/C163258240","wikidata":"https://www.wikidata.org/wiki/Q25342","display_name":"Power (physics)","level":2,"score":0.2879999876022339},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.2838999927043915},{"id":"https://openalex.org/C9390403","wikidata":"https://www.wikidata.org/wiki/Q3966","display_name":"Computer hardware","level":1,"score":0.2754000127315521},{"id":"https://openalex.org/C2742236","wikidata":"https://www.wikidata.org/wiki/Q924713","display_name":"Efficient energy use","level":2,"score":0.2727000117301941},{"id":"https://openalex.org/C13164978","wikidata":"https://www.wikidata.org/wiki/Q600158","display_name":"Hardware acceleration","level":3,"score":0.2680000066757202},{"id":"https://openalex.org/C94375191","wikidata":"https://www.wikidata.org/wiki/Q11205","display_name":"Arithmetic","level":1,"score":0.2669000029563904},{"id":"https://openalex.org/C103275481","wikidata":"https://www.wikidata.org/wiki/Q6787889","display_name":"Matrix representation","level":3,"score":0.26350000500679016},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.2549999952316284}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3787109.3815235","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3787109.3815235","pdf_url":null,"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 Great Lakes Symposium on VLSI 2026","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.1145/3787109.3815235","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3787109.3815235","pdf_url":null,"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 Great Lakes Symposium on VLSI 2026","raw_type":"proceedings-article"},"sustainable_development_goals":[{"score":0.9048623442649841,"id":"https://metadata.un.org/sdg/7","display_name":"Affordable and clean energy"}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":11,"referenced_works":["https://openalex.org/W2946572707","https://openalex.org/W2952857977","https://openalex.org/W2963367920","https://openalex.org/W3035183452","https://openalex.org/W4247198796","https://openalex.org/W4304142274","https://openalex.org/W4390873673","https://openalex.org/W4401753272","https://openalex.org/W4409248731","https://openalex.org/W4409834608","https://openalex.org/W4412722094"],"related_works":[],"abstract_inverted_index":{"As":[0],"deep":[1],"neural":[2],"networks":[3],"grow":[4],"in":[5,94,98],"size,":[6],"their":[7],"computational":[8],"and":[9,24,43,82,96,111,128,136],"energy":[10],"demands":[11],"pose":[12],"significant":[13],"challenges":[14],"for":[15,89],"efficient":[16],"hardware":[17],"acceleration.":[18],"Quantization":[19],"reduces":[20],"storage,":[21],"data":[22],"movement,":[23],"computation":[25],"costs":[26],"by":[27],"lowering":[28],"numerical":[29],"precision.":[30],"While":[31],"8-bit":[32],"integer":[33,84],"(INT8)":[34],"is":[35,103],"widely":[36],"adopted,":[37],"lower":[38,122,131],"precisions":[39],"such":[40],"as":[41],"INT6":[42],"INT4":[44],"can":[45],"also":[46],"maintain":[47],"acceptable":[48],"accuracy":[49],"with":[50,105,134],"specialized":[51],"quantization":[52],"techniques,":[53],"motivating":[54],"accelerators":[55],"that":[56,67,115],"support":[57],"multiple":[58],"precisions.":[59],"This":[60],"paper":[61],"presents":[62],"BitPair,":[63],"a":[64,69],"precision-scalable":[65,108],"accelerator":[66],"employs":[68],"2-bit":[70],"serial":[71],"dataflow":[72],"to":[73,119],"perform":[74],"general":[75],"matrix":[76],"multiplications":[77],"(GEMM).":[78],"BitPair":[79,102,116],"supports":[80],"signed":[81],"unsigned":[83],"operations":[85],"at":[86],"2/4/6/8-bit":[87],"precision":[88],"both":[90],"input":[91],"matrices.":[92],"Implemented":[93],"Verilog":[95],"synthesized":[97],"TSMC":[99],"12nm":[100],"technology,":[101],"compared":[104,133],"two":[106],"prior":[107],"architectures,":[109],"Loom":[110,135],"BitShare.":[112,137],"Evaluation":[113],"shows":[114],"achieves":[117],"up":[118],"1.63":[120],"\u00d7":[121,125,130],"bandwidth,":[123],"1.19":[124],"smaller":[126],"area,":[127],"1.14":[129],"power":[132]},"counts_by_year":[],"updated_date":"2026-07-29T14:22:42.915294","created_date":"2026-06-19T00:00:00"}