{"id":"https://openalex.org/W4416429728","doi":"https://doi.org/10.1109/iccad66269.2025.11240983","title":"SiST: Token Similarity and Sparsity Aware Optimization for Transformers on FPGA","display_name":"SiST: Token Similarity and Sparsity Aware Optimization for Transformers on FPGA","publication_year":2025,"publication_date":"2025-10-26","ids":{"openalex":"https://openalex.org/W4416429728","doi":"https://doi.org/10.1109/iccad66269.2025.11240983"},"language":null,"primary_location":{"id":"doi:10.1109/iccad66269.2025.11240983","is_oa":false,"landing_page_url":"https://doi.org/10.1109/iccad66269.2025.11240983","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 IEEE/ACM International Conference On Computer Aided Design (ICCAD)","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":"https://openalex.org/A5055653143","display_name":"Yifan Zhang","orcid":"https://orcid.org/0000-0003-4207-0632"},"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":"Yifan 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":"https://openalex.org/A5101406825","display_name":"Hongji Wang","orcid":"https://orcid.org/0000-0002-8834-1592"},"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":"Hongji 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/A5049836780","display_name":"Genhao Zhang","orcid":"https://orcid.org/0000-0002-9834-6477"},"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":"Genhao 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":"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/A5112536909","display_name":"Jun Yu","orcid":"https://orcid.org/0009-0006-3920-9321"},"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/A5115695319","display_name":"Kun Wang","orcid":"https://orcid.org/0009-0006-7995-3796"},"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":"9"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.551800012588501,"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"}},"topics":[{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.551800012588501,"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":0.08269999921321869,"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":0.06549999862909317,"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/security-token","display_name":"Security token","score":0.7620999813079834},{"id":"https://openalex.org/keywords/transformer","display_name":"Transformer","score":0.6348999738693237},{"id":"https://openalex.org/keywords/computation","display_name":"Computation","score":0.5799999833106995},{"id":"https://openalex.org/keywords/field-programmable-gate-array","display_name":"Field-programmable gate array","score":0.5687999725341797},{"id":"https://openalex.org/keywords/leverage","display_name":"Leverage (statistics)","score":0.5537999868392944},{"id":"https://openalex.org/keywords/cluster-analysis","display_name":"Cluster analysis","score":0.41269999742507935},{"id":"https://openalex.org/keywords/computational-resource","display_name":"Computational resource","score":0.3758000135421753}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7990000247955322},{"id":"https://openalex.org/C48145219","wikidata":"https://www.wikidata.org/wiki/Q1335365","display_name":"Security token","level":2,"score":0.7620999813079834},{"id":"https://openalex.org/C66322947","wikidata":"https://www.wikidata.org/wiki/Q11658","display_name":"Transformer","level":3,"score":0.6348999738693237},{"id":"https://openalex.org/C45374587","wikidata":"https://www.wikidata.org/wiki/Q12525525","display_name":"Computation","level":2,"score":0.5799999833106995},{"id":"https://openalex.org/C42935608","wikidata":"https://www.wikidata.org/wiki/Q190411","display_name":"Field-programmable gate array","level":2,"score":0.5687999725341797},{"id":"https://openalex.org/C153083717","wikidata":"https://www.wikidata.org/wiki/Q6535263","display_name":"Leverage (statistics)","level":2,"score":0.5537999868392944},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.41769999265670776},{"id":"https://openalex.org/C73555534","wikidata":"https://www.wikidata.org/wiki/Q622825","display_name":"Cluster analysis","level":2,"score":0.41269999742507935},{"id":"https://openalex.org/C127964446","wikidata":"https://www.wikidata.org/wiki/Q1092142","display_name":"Computational resource","level":3,"score":0.3758000135421753},{"id":"https://openalex.org/C2779960059","wikidata":"https://www.wikidata.org/wiki/Q7113681","display_name":"Overhead (engineering)","level":2,"score":0.3465999960899353},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.33309999108314514},{"id":"https://openalex.org/C179799912","wikidata":"https://www.wikidata.org/wiki/Q205084","display_name":"Computational complexity theory","level":2,"score":0.32409998774528503},{"id":"https://openalex.org/C2777480716","wikidata":"https://www.wikidata.org/wiki/Q23582796","display_name":"Resource consumption","level":2,"score":0.32019999623298645},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.31439998745918274},{"id":"https://openalex.org/C149635348","wikidata":"https://www.wikidata.org/wiki/Q193040","display_name":"Embedded system","level":1,"score":0.31279999017715454},{"id":"https://openalex.org/C66024118","wikidata":"https://www.wikidata.org/wiki/Q1122506","display_name":"Computational model","level":2,"score":0.28870001435279846},{"id":"https://openalex.org/C9390403","wikidata":"https://www.wikidata.org/wiki/Q3966","display_name":"Computer hardware","level":1,"score":0.27250000834465027},{"id":"https://openalex.org/C2781357197","wikidata":"https://www.wikidata.org/wiki/Q5757597","display_name":"High memory","level":2,"score":0.27140000462532043},{"id":"https://openalex.org/C138236772","wikidata":"https://www.wikidata.org/wiki/Q25098575","display_name":"Edge device","level":3,"score":0.26579999923706055},{"id":"https://openalex.org/C2742236","wikidata":"https://www.wikidata.org/wiki/Q924713","display_name":"Efficient energy use","level":2,"score":0.2533999979496002}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/iccad66269.2025.11240983","is_oa":false,"landing_page_url":"https://doi.org/10.1109/iccad66269.2025.11240983","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 IEEE/ACM International Conference On Computer Aided Design (ICCAD)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[{"id":"https://openalex.org/F4320321851","display_name":"Fudan University","ror":"https://ror.org/013q1eq08"}],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":21,"referenced_works":["https://openalex.org/W2073849744","https://openalex.org/W2108598243","https://openalex.org/W2923014074","https://openalex.org/W2949320316","https://openalex.org/W2963748441","https://openalex.org/W2979826702","https://openalex.org/W3047848469","https://openalex.org/W3130240120","https://openalex.org/W3132616766","https://openalex.org/W3184454880","https://openalex.org/W3205192296","https://openalex.org/W3206837665","https://openalex.org/W4214686755","https://openalex.org/W4293025165","https://openalex.org/W4375928954","https://openalex.org/W4380881077","https://openalex.org/W4385245566","https://openalex.org/W4388214800","https://openalex.org/W4389166707","https://openalex.org/W4390017976","https://openalex.org/W4404133735"],"related_works":[],"abstract_inverted_index":{"Transformers":[0],"have":[1],"increasingly":[2],"become":[3],"the":[4,35,56,63,67,71,151,162,172],"backbone":[5],"of":[6,94,153],"modern":[7],"AI,":[8],"excelling":[9],"in":[10,40,75,117,143,155,203,223],"tasks":[11],"across":[12,185],"natural":[13],"language":[14],"processing":[15],"(NLP)":[16],"and":[17,66,99,121,175,218],"computer":[18],"vision":[19],"(CV).":[20],"Despite":[21],"their":[22],"effectiveness,":[23],"calculating":[24],"them":[25],"is":[26,61,157],"costly,":[27],"especially":[28],"on":[29,55],"source-constrained":[30],"platforms":[31],"like":[32],"FPGAs.":[33],"Among":[34],"three":[36,92],"main":[37],"computation":[38],"modules":[39,93],"a":[41,105,207],"Transformer":[42,95,230],"model":[43],"(QKV":[44],"generation,":[45],"attention":[46,57,122],"computation,":[47],"FFN),":[48],"most":[49,76],"prior":[50],"research":[51],"has":[52],"primarily":[53],"focused":[54],"computation.":[58],"However,":[59],"it":[60],"typically":[62],"QKV":[64,111,119],"generation":[65,112,120],"FFN":[68,144,156],"that":[69,189],"dominate":[70],"overall":[72],"computational":[73,148,163,192],"cost":[74],"cases.":[77],"To":[78],"address":[79],"this":[80],"gap,":[81],"we":[82,135],"propose":[83,104],"SiST,":[84],"an":[85],"innovative":[86],"algorithm-architecture":[87],"co-design":[88],"to":[89,113,139,159,196,209,213,220],"optimize":[90,161],"all":[91],"with":[96,198,227],"both":[97,118],"sparsity":[98,107,152,174],"token":[100,137,141,176],"similarity.":[101],"We":[102],"first":[103],"cross-stage":[106],"prediction":[108,125],"mechanism":[109,126],"before":[110],"minimize":[114],"computations":[115],"involved":[116],"computing.":[123],"The":[124],"avoids":[127],"large":[128],"overhead":[129,193],"through":[130],"off-line":[131],"weight":[132],"clustering.":[133],"Further,":[134],"leverage":[136],"similarity":[138],"adapt":[140],"clustering":[142],"module,":[145],"thereby":[146],"reducing":[147],"load.":[149,164],"Moreover,":[150],"weights":[154],"explored":[158],"further":[160],"Our":[165],"dedicated":[166],"hardware":[167,180],"architecture":[168],"can":[169],"effectively":[170],"handle":[171],"unstructured":[173],"clustering,":[177],"ensuring":[178],"high":[179],"resource":[181],"utilization.":[182],"Comprehensive":[183],"evaluations":[184],"multiple":[186],"benchmarks":[187],"demonstrate":[188],"SiST":[190],"reduces":[191],"by":[194],"45.9%":[195],"52.2%":[197],"less":[199],"than":[200],"2%":[201],"loss":[202],"accuracy.":[204],"It":[205],"achieves":[206],"2.35\u00d7":[208],"8.44\u00d7":[210],"speedup,":[211],"1.50\u00d7":[212],"14.09\u00d7":[214],"normalized":[215],"throughput":[216],"improvement":[217,222],"2.43\u00d7":[219],"3.44\u00d7":[221],"energy":[224],"efficiency":[225],"compared":[226],"other":[228],"FPGA-based":[229],"accelerators.":[231]},"counts_by_year":[],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-11-20T00:00:00"}