{"id":"https://openalex.org/W4404133526","doi":"https://doi.org/10.1145/3649329.3657325","title":"CSTrans-OPU: An FPGA-based Overlay Processor with Full Compilation for Transformer Networks via Sparsity Exploration","display_name":"CSTrans-OPU: An FPGA-based Overlay Processor with Full Compilation for Transformer Networks via Sparsity Exploration","publication_year":2024,"publication_date":"2024-06-23","ids":{"openalex":"https://openalex.org/W4404133526","doi":"https://doi.org/10.1145/3649329.3657325"},"language":"en","primary_location":{"id":"doi:10.1145/3649329.3657325","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3649329.3657325","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3649329.3657325?download=true","source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 61st ACM/IEEE Design Automation Conference","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://dl.acm.org/doi/pdf/10.1145/3649329.3657325?download=true","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5100580160","display_name":"Yueyin Bai","orcid":"https://orcid.org/0009-0006-2938-5153"},"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":"Yueyin Bai","raw_affiliation_strings":["Fudan University, ShangHai, ShangHai, China"],"raw_orcid":"https://orcid.org/0009-0006-2938-5153","affiliations":[{"raw_affiliation_string":"Fudan University, ShangHai, ShangHai, China","institution_ids":["https://openalex.org/I24943067"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5027433777","display_name":"Keqing Zhao","orcid":"https://orcid.org/0009-0001-4951-4059"},"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":"Keqing Zhao","raw_affiliation_strings":["Fudan University, ShangHai, ShangHai, China"],"raw_orcid":"https://orcid.org/0009-0001-4951-4059","affiliations":[{"raw_affiliation_string":"Fudan University, ShangHai, ShangHai, China","institution_ids":["https://openalex.org/I24943067"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5115604721","display_name":"Yang Liu","orcid":"https://orcid.org/0009-0001-0911-0666"},"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":"Yang Liu","raw_affiliation_strings":["Fudan University, ShangHai, ShangHai, China"],"raw_orcid":"https://orcid.org/0009-0001-0911-0666","affiliations":[{"raw_affiliation_string":"Fudan University, ShangHai, ShangHai, China","institution_ids":["https://openalex.org/I24943067"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5115604891","display_name":"Hongji Wang","orcid":"https://orcid.org/0009-0006-8758-7239"},"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":"Hongji Wang","raw_affiliation_strings":["Fudan University, ShangHai, ShangHai, China"],"raw_orcid":"https://orcid.org/0009-0006-8758-7239","affiliations":[{"raw_affiliation_string":"Fudan University, ShangHai, ShangHai, China","institution_ids":["https://openalex.org/I24943067"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5102185842","display_name":"Hao Zhou","orcid":"https://orcid.org/0009-0003-8421-1242"},"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":"Hao Zhou","raw_affiliation_strings":["Fudan University, ShangHai, ShangHai, China"],"raw_orcid":"https://orcid.org/0009-0003-8421-1242","affiliations":[{"raw_affiliation_string":"Fudan University, ShangHai, ShangHai, China","institution_ids":["https://openalex.org/I24943067"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5000233105","display_name":"Xunyi Wu","orcid":"https://orcid.org/0009-0008-8125-6978"},"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":"Xiaoxing Wu","raw_affiliation_strings":["Fudan University, ShangHai, ShangHai, China"],"raw_orcid":"https://orcid.org/0009-0008-8125-6978","affiliations":[{"raw_affiliation_string":"Fudan University, ShangHai, ShangHai, China","institution_ids":["https://openalex.org/I24943067"]}]},{"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/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 Yu","raw_affiliation_strings":["Fudan University, ShangHai, ShangHai, China"],"raw_orcid":"https://orcid.org/0000-0003-4286-9292","affiliations":[{"raw_affiliation_string":"Fudan University, ShangHai, ShangHai, China","institution_ids":["https://openalex.org/I24943067"]}]},{"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/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":"Kun Wang","raw_affiliation_strings":["Fudan University, ShangHai, ShangHai, China"],"raw_orcid":"https://orcid.org/0000-0002-7288-1789","affiliations":[{"raw_affiliation_string":"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":null,"has_fulltext":true,"cited_by_count":4,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"6"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11338","display_name":"Advancements in Photolithography Techniques","score":0.9944999814033508,"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"}},"topics":[{"id":"https://openalex.org/T11338","display_name":"Advancements in Photolithography Techniques","score":0.9944999814033508,"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/T14117","display_name":"Integrated Circuits and Semiconductor Failure Analysis","score":0.9853000044822693,"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/T10624","display_name":"Silicon and Solar Cell Technologies","score":0.9825999736785889,"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/field-programmable-gate-array","display_name":"Field-programmable gate array","score":0.7821862697601318},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.7700345516204834},{"id":"https://openalex.org/keywords/overlay","display_name":"Overlay","score":0.7614903450012207},{"id":"https://openalex.org/keywords/transformer","display_name":"Transformer","score":0.4466104507446289},{"id":"https://openalex.org/keywords/embedded-system","display_name":"Embedded system","score":0.4142690598964691},{"id":"https://openalex.org/keywords/computer-architecture","display_name":"Computer architecture","score":0.36735695600509644},{"id":"https://openalex.org/keywords/operating-system","display_name":"Operating system","score":0.19811144471168518},{"id":"https://openalex.org/keywords/engineering","display_name":"Engineering","score":0.08980381488800049},{"id":"https://openalex.org/keywords/electrical-engineering","display_name":"Electrical engineering","score":0.0883806049823761}],"concepts":[{"id":"https://openalex.org/C42935608","wikidata":"https://www.wikidata.org/wiki/Q190411","display_name":"Field-programmable gate array","level":2,"score":0.7821862697601318},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7700345516204834},{"id":"https://openalex.org/C136085584","wikidata":"https://www.wikidata.org/wiki/Q910289","display_name":"Overlay","level":2,"score":0.7614903450012207},{"id":"https://openalex.org/C66322947","wikidata":"https://www.wikidata.org/wiki/Q11658","display_name":"Transformer","level":3,"score":0.4466104507446289},{"id":"https://openalex.org/C149635348","wikidata":"https://www.wikidata.org/wiki/Q193040","display_name":"Embedded system","level":1,"score":0.4142690598964691},{"id":"https://openalex.org/C118524514","wikidata":"https://www.wikidata.org/wiki/Q173212","display_name":"Computer architecture","level":1,"score":0.36735695600509644},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.19811144471168518},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.08980381488800049},{"id":"https://openalex.org/C119599485","wikidata":"https://www.wikidata.org/wiki/Q43035","display_name":"Electrical engineering","level":1,"score":0.0883806049823761},{"id":"https://openalex.org/C165801399","wikidata":"https://www.wikidata.org/wiki/Q25428","display_name":"Voltage","level":2,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3649329.3657325","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3649329.3657325","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3649329.3657325?download=true","source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 61st ACM/IEEE Design Automation Conference","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.1145/3649329.3657325","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3649329.3657325","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3649329.3657325?download=true","source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 61st ACM/IEEE Design Automation Conference","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G6507269736","display_name":null,"funder_award_id":"2021YFA1003602","funder_id":"https://openalex.org/F4320335777","funder_display_name":"National Key Research and Development Program of China"}],"funders":[{"id":"https://openalex.org/F4320335777","display_name":"National Key Research and Development Program of China","ror":null}],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4404133526.pdf","grobid_xml":"https://content.openalex.org/works/W4404133526.grobid-xml"},"referenced_works_count":9,"referenced_works":["https://openalex.org/W2949320316","https://openalex.org/W2979455536","https://openalex.org/W3047848469","https://openalex.org/W3159727696","https://openalex.org/W3184454880","https://openalex.org/W4293023328","https://openalex.org/W4293025109","https://openalex.org/W4385187240","https://openalex.org/W6759363029"],"related_works":["https://openalex.org/W595346907","https://openalex.org/W153296825","https://openalex.org/W598989511","https://openalex.org/W2096844293","https://openalex.org/W2363944576","https://openalex.org/W2351041855","https://openalex.org/W2570254841","https://openalex.org/W1967938402","https://openalex.org/W2386041993","https://openalex.org/W1608572506"],"abstract_inverted_index":{"A":[0],"few":[1],"overlay":[2,55,196],"processors":[3,17],"for":[4,37,60,77,198],"transformer":[5,61,199],"networks":[6,62,200],"emerge":[7],"to":[8,97,156,174],"achieve":[9],"reconfigurable":[10],"architectures":[11],"and":[12,42,89,118,132,140,160,170,177],"dynamic":[13],"instructions.":[14],"However,":[15],"these":[16],"consistently":[18],"neglect":[19],"exploring":[20],"network":[21],"sparsity,":[22],"while":[23],"existing":[24],"sparse":[25],"accelerators":[26],"inefficiently":[27],"utilize":[28],"resources":[29],"with":[30,57,75,81,104,138,144,150,166],"separate":[31],"computation":[32,79,90],"parts.":[33],"Furthermore,":[34,152],"mainstream":[35],"compilers":[36],"instruction":[38,116],"generation":[39,117],"are":[40],"intricate":[41],"demand":[43],"significant":[44],"engineering":[45],"efforts.":[46],"In":[47],"this":[48],"work,":[49],"we":[50,67,153],"propose":[51],"CSTrans-OPU,":[52],"an":[53],"FPGA-based":[54],"processor":[56,197],"full":[58,82],"compilation":[59],"via":[63],"sparsity":[64],"exploration.":[65],"Specifically,":[66],"customize":[68],"a":[69,105],"multi-precision":[70],"processing":[71],"element":[72],"(PE)":[73],"array":[74],"DSP-packing":[76],"unified":[78],"format":[80],"resource":[83],"utilization.":[84],"Additionally,":[85],"the":[86,99,186,194],"introduced":[87],"sorting":[88],"mode":[91],"selection":[92],"modules":[93],"make":[94],"it":[95],"possible":[96],"explore":[98],"token":[100],"sparsity.":[101,202],"Moreover,":[102],"equipped":[103],"user-friendly":[106],"compiler,":[107],"CSTrans-OPU":[108,128,192],"enables":[109],"model":[110,114,122],"parsing,":[111],"operation":[112],"fusion,":[113],"quantization,":[115],"reordering":[119],"directly":[120],"from":[121],"files.":[123],"Experimental":[124],"results":[125],"show":[126],"that":[127],"achieves":[129],"6.92-20.06\u00d7":[130],"speedup":[131],"182.48\u00d7":[133],"higher":[134,162],"energy":[135,147,163,180],"efficiency":[136,148,164],"compared":[137,149,165],"CPU,":[139],"1.47-3.85\u00d7":[141],"latency":[142,159],"reduction":[143],"4.63-52.53\u00d7":[145],"better":[146,158],"GPU.":[151],"observe":[154],"up":[155,173],"4.28\u00d7":[157],"4.94\u00d7":[161],"previously":[167],"customized":[168],"accelerators,":[169],"can":[171],"be":[172],"1.93\u00d7":[175],"faster":[176],"4.39\u00d7":[178],"more":[179],"efficient":[181],"than":[182],"FPGA":[183],"processors.":[184],"To":[185],"best":[187],"of":[188],"our":[189,191],"knowledge,":[190],"is":[193],"first":[195],"considering":[201]},"counts_by_year":[{"year":2025,"cited_by_count":4}],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-10-10T00:00:00"}
