{"id":"https://openalex.org/W4414198425","doi":"https://doi.org/10.1109/dac63849.2025.11132475","title":"Blaze: An Efficient Bit-Sparse Attention Architecture With Workload Orchestration Optimization","display_name":"Blaze: An Efficient Bit-Sparse Attention Architecture With Workload Orchestration Optimization","publication_year":2025,"publication_date":"2025-06-22","ids":{"openalex":"https://openalex.org/W4414198425","doi":"https://doi.org/10.1109/dac63849.2025.11132475"},"language":"en","primary_location":{"id":"doi:10.1109/dac63849.2025.11132475","is_oa":false,"landing_page_url":"https://doi.org/10.1109/dac63849.2025.11132475","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 62nd ACM/IEEE Design Automation Conference (DAC)","raw_type":"proceedings-article"},"type":"conference-paper","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":null,"display_name":"Runzhou Zhang","orcid":null},"institutions":[{"id":"https://openalex.org/I4210132426","display_name":"Shanghai Fudan Microelectronics (China)","ror":"https://ror.org/02vfj3j86","country_code":"CN","type":"company","lineage":["https://openalex.org/I4210132426"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Runzhou Zhang","raw_affiliation_strings":["Fudan University,State Key Lab of Integrated Chips & Systems, and School of Microelectronics,Shanghai,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Fudan University,State Key Lab of Integrated Chips & Systems, and School of Microelectronics,Shanghai,China","institution_ids":["https://openalex.org/I4210132426"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Faxian Sun","orcid":null},"institutions":[{"id":"https://openalex.org/I4210132426","display_name":"Shanghai Fudan Microelectronics (China)","ror":"https://ror.org/02vfj3j86","country_code":"CN","type":"company","lineage":["https://openalex.org/I4210132426"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Faxian Sun","raw_affiliation_strings":["Fudan University,State Key Lab of Integrated Chips & Systems, and School of Microelectronics,Shanghai,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Fudan University,State Key Lab of Integrated Chips & Systems, and School of Microelectronics,Shanghai,China","institution_ids":["https://openalex.org/I4210132426"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100377935","display_name":"Yiming Wang","orcid":"https://orcid.org/0000-0002-5932-4371"},"institutions":[{"id":"https://openalex.org/I4210132426","display_name":"Shanghai Fudan Microelectronics (China)","ror":"https://ror.org/02vfj3j86","country_code":"CN","type":"company","lineage":["https://openalex.org/I4210132426"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yiming Wang","raw_affiliation_strings":["Fudan University,State Key Lab of Integrated Chips & Systems, and School of Microelectronics,Shanghai,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Fudan University,State Key Lab of Integrated Chips & Systems, and School of Microelectronics,Shanghai,China","institution_ids":["https://openalex.org/I4210132426"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101570582","display_name":"Kun Zou","orcid":"https://orcid.org/0000-0001-5182-6855"},"institutions":[{"id":"https://openalex.org/I4210132426","display_name":"Shanghai Fudan Microelectronics (China)","ror":"https://ror.org/02vfj3j86","country_code":"CN","type":"company","lineage":["https://openalex.org/I4210132426"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Kunchen Zou","raw_affiliation_strings":["Fudan University,State Key Lab of Integrated Chips & Systems, and School of Microelectronics,Shanghai,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Fudan University,State Key Lab of Integrated Chips & Systems, and School of Microelectronics,Shanghai,China","institution_ids":["https://openalex.org/I4210132426"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101755853","display_name":"Zhihao Qin","orcid":"https://orcid.org/0000-0002-5577-1270"},"institutions":[{"id":"https://openalex.org/I4210132426","display_name":"Shanghai Fudan Microelectronics (China)","ror":"https://ror.org/02vfj3j86","country_code":"CN","type":"company","lineage":["https://openalex.org/I4210132426"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zhinan Qin","raw_affiliation_strings":["Fudan University,State Key Lab of Integrated Chips & Systems, and School of Microelectronics,Shanghai,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Fudan University,State Key Lab of Integrated Chips & Systems, and School of Microelectronics,Shanghai,China","institution_ids":["https://openalex.org/I4210132426"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5045009372","display_name":"Jianli Chen","orcid":"https://orcid.org/0000-0001-5405-8441"},"institutions":[{"id":"https://openalex.org/I4210132426","display_name":"Shanghai Fudan Microelectronics (China)","ror":"https://ror.org/02vfj3j86","country_code":"CN","type":"company","lineage":["https://openalex.org/I4210132426"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Jianli Chen","raw_affiliation_strings":["Fudan University,State Key Lab of Integrated Chips & Systems, and School of Microelectronics,Shanghai,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Fudan University,State Key Lab of Integrated Chips & Systems, and School of Microelectronics,Shanghai,China","institution_ids":["https://openalex.org/I4210132426"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5103098516","display_name":"Jun Yu","orcid":"https://orcid.org/0000-0003-4286-9292"},"institutions":[{"id":"https://openalex.org/I4210132426","display_name":"Shanghai Fudan Microelectronics (China)","ror":"https://ror.org/02vfj3j86","country_code":"CN","type":"company","lineage":["https://openalex.org/I4210132426"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Jun Yu","raw_affiliation_strings":["Fudan University,State Key Lab of Integrated Chips & Systems, and School of Microelectronics,Shanghai,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Fudan University,State Key Lab of Integrated Chips & Systems, and School of Microelectronics,Shanghai,China","institution_ids":["https://openalex.org/I4210132426"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5015932760","display_name":"Kun Wang","orcid":"https://orcid.org/0000-0002-7288-1789"},"institutions":[{"id":"https://openalex.org/I4210132426","display_name":"Shanghai Fudan Microelectronics (China)","ror":"https://ror.org/02vfj3j86","country_code":"CN","type":"company","lineage":["https://openalex.org/I4210132426"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Kun Wang","raw_affiliation_strings":["Fudan University,State Key Lab of Integrated Chips & Systems, and School of Microelectronics,Shanghai,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Fudan University,State Key Lab of Integrated Chips & Systems, and School of Microelectronics,Shanghai,China","institution_ids":["https://openalex.org/I4210132426"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I4210132426"],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"7"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.9994000196456909,"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"}},"topics":[{"id":"https://openalex.org/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.9994000196456909,"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"}},{"id":"https://openalex.org/T10904","display_name":"Embedded Systems Design Techniques","score":0.9934999942779541,"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"}},{"id":"https://openalex.org/T11992","display_name":"CCD and CMOS Imaging Sensors","score":0.9819999933242798,"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/workload","display_name":"Workload","score":0.7081999778747559},{"id":"https://openalex.org/keywords/exploit","display_name":"Exploit","score":0.6754000186920166},{"id":"https://openalex.org/keywords/speedup","display_name":"Speedup","score":0.5034999847412109},{"id":"https://openalex.org/keywords/efficient-energy-use","display_name":"Efficient energy use","score":0.4803999960422516},{"id":"https://openalex.org/keywords/redundancy","display_name":"Redundancy (engineering)","score":0.4348999857902527},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.37869998812675476},{"id":"https://openalex.org/keywords/energy-consumption","display_name":"Energy consumption","score":0.3767000138759613},{"id":"https://openalex.org/keywords/orchestration","display_name":"Orchestration","score":0.37450000643730164},{"id":"https://openalex.org/keywords/architecture","display_name":"Architecture","score":0.37119999527931213}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8174999952316284},{"id":"https://openalex.org/C2778476105","wikidata":"https://www.wikidata.org/wiki/Q628539","display_name":"Workload","level":2,"score":0.7081999778747559},{"id":"https://openalex.org/C165696696","wikidata":"https://www.wikidata.org/wiki/Q11287","display_name":"Exploit","level":2,"score":0.6754000186920166},{"id":"https://openalex.org/C68339613","wikidata":"https://www.wikidata.org/wiki/Q1549489","display_name":"Speedup","level":2,"score":0.5034999847412109},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.49149999022483826},{"id":"https://openalex.org/C2742236","wikidata":"https://www.wikidata.org/wiki/Q924713","display_name":"Efficient energy use","level":2,"score":0.4803999960422516},{"id":"https://openalex.org/C152124472","wikidata":"https://www.wikidata.org/wiki/Q1204361","display_name":"Redundancy (engineering)","level":2,"score":0.4348999857902527},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.37869998812675476},{"id":"https://openalex.org/C2780165032","wikidata":"https://www.wikidata.org/wiki/Q16869822","display_name":"Energy consumption","level":2,"score":0.3767000138759613},{"id":"https://openalex.org/C199168358","wikidata":"https://www.wikidata.org/wiki/Q3367000","display_name":"Orchestration","level":3,"score":0.37450000643730164},{"id":"https://openalex.org/C123657996","wikidata":"https://www.wikidata.org/wiki/Q12271","display_name":"Architecture","level":2,"score":0.37119999527931213},{"id":"https://openalex.org/C81363708","wikidata":"https://www.wikidata.org/wiki/Q17084460","display_name":"Convolutional neural network","level":2,"score":0.36880001425743103},{"id":"https://openalex.org/C118524514","wikidata":"https://www.wikidata.org/wiki/Q173212","display_name":"Computer architecture","level":1,"score":0.35190001130104065},{"id":"https://openalex.org/C5119721","wikidata":"https://www.wikidata.org/wiki/Q220501","display_name":"Quality of service","level":2,"score":0.3474000096321106},{"id":"https://openalex.org/C193415008","wikidata":"https://www.wikidata.org/wiki/Q639681","display_name":"Network architecture","level":2,"score":0.32010000944137573},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.3102000057697296},{"id":"https://openalex.org/C192209626","wikidata":"https://www.wikidata.org/wiki/Q190909","display_name":"Focus (optics)","level":2,"score":0.30730000138282776},{"id":"https://openalex.org/C105339364","wikidata":"https://www.wikidata.org/wiki/Q2297740","display_name":"Software deployment","level":2,"score":0.30169999599456787},{"id":"https://openalex.org/C137836250","wikidata":"https://www.wikidata.org/wiki/Q984063","display_name":"Optimization problem","level":2,"score":0.30090001225471497},{"id":"https://openalex.org/C2780149590","wikidata":"https://www.wikidata.org/wiki/Q7302742","display_name":"Reconfigurability","level":2,"score":0.30070000886917114},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.28600001335144043},{"id":"https://openalex.org/C78766204","wikidata":"https://www.wikidata.org/wiki/Q555032","display_name":"Multi-core processor","level":2,"score":0.28540000319480896},{"id":"https://openalex.org/C79974875","wikidata":"https://www.wikidata.org/wiki/Q483639","display_name":"Cloud computing","level":2,"score":0.2808000147342682},{"id":"https://openalex.org/C75684735","wikidata":"https://www.wikidata.org/wiki/Q858810","display_name":"Big data","level":2,"score":0.27709999680519104},{"id":"https://openalex.org/C2776221188","wikidata":"https://www.wikidata.org/wiki/Q21072556","display_name":"Design space exploration","level":2,"score":0.2705000042915344},{"id":"https://openalex.org/C170858558","wikidata":"https://www.wikidata.org/wiki/Q1394144","display_name":"Automatic summarization","level":2,"score":0.259799987077713},{"id":"https://openalex.org/C149635348","wikidata":"https://www.wikidata.org/wiki/Q193040","display_name":"Embedded system","level":1,"score":0.25850000977516174},{"id":"https://openalex.org/C2778915421","wikidata":"https://www.wikidata.org/wiki/Q3643177","display_name":"Performance improvement","level":2,"score":0.25040000677108765}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/dac63849.2025.11132475","is_oa":false,"landing_page_url":"https://doi.org/10.1109/dac63849.2025.11132475","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 62nd ACM/IEEE Design Automation Conference (DAC)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[{"id":"https://openalex.org/F4320321001","display_name":"National Natural Science Foundation of China","ror":"https://ror.org/01h0zpd94"},{"id":"https://openalex.org/F4320337504","display_name":"Research and Development","ror":"https://ror.org/027s68j25"}],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":16,"referenced_works":["https://openalex.org/W2725159389","https://openalex.org/W2889048825","https://openalex.org/W2923014074","https://openalex.org/W2963748441","https://openalex.org/W2979439447","https://openalex.org/W2979826702","https://openalex.org/W3094502228","https://openalex.org/W3159727696","https://openalex.org/W3206453033","https://openalex.org/W4210396250","https://openalex.org/W4280582299","https://openalex.org/W4285236966","https://openalex.org/W4386765191","https://openalex.org/W4387789741","https://openalex.org/W4390097918","https://openalex.org/W4400726540"],"related_works":[],"abstract_inverted_index":{"The":[0],"attention":[1,69,89,123,168,181,202,236],"mechanism":[2,174],"is":[3],"a":[4,172,186],"core":[5],"neural":[6],"network":[7],"primitive":[8],"widely":[9],"utilized":[10],"in":[11,58,112,132,166,214,223,235],"state-of-the-art":[12,200],"models":[13,60],"of":[14,54,68,109,180],"Natural":[15],"Language":[16],"Processing":[17],"(NLP)":[18],"applications.":[19],"However,":[20],"the":[21,52,65,88,97,106,142,164,167,178,241],"high":[22],"computational":[23],"complexity":[24],"and":[25,32,43,91,100,129,216],"substantial":[26],"power":[27],"consumption":[28],"hinder":[29],"its":[30],"deployment":[31],"efficient":[33,122],"inference.":[34,70],"To":[35,140],"address":[36],"these":[37,50],"challenges,":[38],"various":[39],"methods":[40,74],"leveraging":[41],"sparsity":[42,57,131],"quantization":[44],"have":[45,93],"been":[46],"proposed.":[47],"Compared":[48],"to":[49,161,175,190,199,231],"methods,":[51],"exploitation":[53],"abundant":[55],"bit-level":[56,130],"attention-based":[59],"presents":[61],"great":[62],"potential":[63],"for":[64,87],"performance":[66,179,215,238],"enhancement":[67,222],"Existing":[71],"bit-sparsity":[72],"optimization":[73],"primarily":[75],"focus":[76],"on":[77],"Convolutional":[78],"Neural":[79],"Networks":[80],"(CNNs),":[81],"which":[82],"are":[83],"not":[84,94],"ideally":[85],"suitable":[86],"mechanism,":[90,169],"they":[92],"effectively":[95],"solved":[96],"workload":[98,137,143,157],"imbalance":[99,144],"hardware":[101],"under-utilization":[102],"issues":[103,145],"caused":[104],"by":[105,148],"irregular":[107],"distribution":[108],"non-zero":[110],"bits":[111],"tensor":[113,133],"data.":[114],"In":[115],"this":[116],"work,":[117],"we":[118,152,170],"introduce":[119],"Blaze,":[120],"an":[121,154],"architecture":[124],"that":[125],"leverages":[126],"both":[127,192],"value":[128],"data":[134],"along":[135],"with":[136,240],"orchestration":[138,158],"optimization.":[139],"mitigate":[141],"often":[146],"encountered":[147],"sparse":[149],"bit-serial":[150],"architecture,":[151],"propose":[153,171],"Approximate-Computing-Based":[155],"(ACB)":[156],"mechanism.":[159],"Additionally,":[160],"fully":[162],"exploit":[163],"redundancy":[165],"Leading-Booth":[173],"further":[176],"enhance":[177],"computation.":[182],"We":[183],"also":[184],"design":[185],"reconfigurable":[187],"computing":[188,237],"engine":[189],"support":[191],"mechanisms.":[193],"Experimental":[194],"results":[195],"indicate":[196],"that,":[197],"compared":[198,239],"(SOTA)":[201],"accelerators,":[203],"our":[204],"Blaze":[205],"can":[206,228],"achieve":[207],"$2.37":[208],"\\times":[209,218],"\\sim":[210,219],"6.18":[211],"\\times$":[212,221,233],"improvement":[213],"$9.69":[217],"43.96":[220],"energy":[224],"efficiency.":[225],"Our":[226],"accelerator":[227],"reach":[229],"up":[230],"$1.58":[232],"speedup":[234],"SOTA":[242],"bit-sparse":[243],"accelerator.":[244]},"counts_by_year":[],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-10-10T00:00:00"}