{"id":"https://openalex.org/W4409285558","doi":"https://doi.org/10.1145/3676536.3676667","title":"Edge-BiT: Software-Hardware Co-design for Optimizing Binarized Transformer Networks Inference on Edge FPGA","display_name":"Edge-BiT: Software-Hardware Co-design for Optimizing Binarized Transformer Networks Inference on Edge FPGA","publication_year":2024,"publication_date":"2024-10-27","ids":{"openalex":"https://openalex.org/W4409285558","doi":"https://doi.org/10.1145/3676536.3676667"},"language":"en","primary_location":{"id":"doi:10.1145/3676536.3676667","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3676536.3676667","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3676536.3676667","source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 43rd IEEE/ACM International Conference on Computer-Aided Design","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/10.1145/3676536.3676667","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5101954794","display_name":"Shuai Zhou","orcid":"https://orcid.org/0009-0007-1531-3447"},"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":true,"raw_author_name":"Shuai Zhou","raw_affiliation_strings":["Fudan University, Shang Hai, China"],"affiliations":[{"raw_affiliation_string":"Fudan University, Shang Hai, China","institution_ids":["https://openalex.org/I24943067"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5117084707","display_name":"Sisi Meng","orcid":null},"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":"Sisi Meng","raw_affiliation_strings":["Fudan University, Shanghai, China"],"affiliations":[{"raw_affiliation_string":"Fudan University, Shanghai, China","institution_ids":["https://openalex.org/I24943067"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5022672817","display_name":"Huinan Tian","orcid":"https://orcid.org/0009-0002-3887-6852"},"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":"Huinan Tian","raw_affiliation_strings":["Fudan University, Shanghai, China"],"affiliations":[{"raw_affiliation_string":"Fudan University, 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, China"],"affiliations":[{"raw_affiliation_string":"Fudan University, 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, China"],"affiliations":[{"raw_affiliation_string":"Fudan University, Shanghai, China","institution_ids":["https://openalex.org/I24943067"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":5,"corresponding_author_ids":["https://openalex.org/A5101954794"],"corresponding_institution_ids":["https://openalex.org/I24943067"],"apc_list":null,"apc_paid":null,"fwci":0.4459,"has_fulltext":true,"cited_by_count":2,"citation_normalized_percentile":{"value":0.66082907,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":95,"max":96},"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/T14117","display_name":"Integrated Circuits and Semiconductor Failure Analysis","score":0.9925000071525574,"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/T14117","display_name":"Integrated Circuits and Semiconductor Failure Analysis","score":0.9925000071525574,"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/T11032","display_name":"VLSI and Analog Circuit Testing","score":0.9918000102043152,"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/T12111","display_name":"Industrial Vision Systems and Defect Detection","score":0.9908999800682068,"subfield":{"id":"https://openalex.org/subfields/2209","display_name":"Industrial and Manufacturing 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":0.7738999128341675},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.7372133731842041},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.617192268371582},{"id":"https://openalex.org/keywords/enhanced-data-rates-for-gsm-evolution","display_name":"Enhanced Data Rates for GSM Evolution","score":0.5904328227043152},{"id":"https://openalex.org/keywords/software","display_name":"Software","score":0.5748187899589539},{"id":"https://openalex.org/keywords/computer-hardware","display_name":"Computer hardware","score":0.5511795282363892},{"id":"https://openalex.org/keywords/computer-architecture","display_name":"Computer architecture","score":0.4911019802093506},{"id":"https://openalex.org/keywords/software-implementation","display_name":"Software implementation","score":0.4328320622444153},{"id":"https://openalex.org/keywords/embedded-system","display_name":"Embedded system","score":0.429738312959671},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.2517184615135193},{"id":"https://openalex.org/keywords/programming-language","display_name":"Programming language","score":0.12478438019752502}],"concepts":[{"id":"https://openalex.org/C42935608","wikidata":"https://www.wikidata.org/wiki/Q190411","display_name":"Field-programmable gate array","level":2,"score":0.7738999128341675},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7372133731842041},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.617192268371582},{"id":"https://openalex.org/C162307627","wikidata":"https://www.wikidata.org/wiki/Q204833","display_name":"Enhanced Data Rates for GSM Evolution","level":2,"score":0.5904328227043152},{"id":"https://openalex.org/C2777904410","wikidata":"https://www.wikidata.org/wiki/Q7397","display_name":"Software","level":2,"score":0.5748187899589539},{"id":"https://openalex.org/C9390403","wikidata":"https://www.wikidata.org/wiki/Q3966","display_name":"Computer hardware","level":1,"score":0.5511795282363892},{"id":"https://openalex.org/C118524514","wikidata":"https://www.wikidata.org/wiki/Q173212","display_name":"Computer architecture","level":1,"score":0.4911019802093506},{"id":"https://openalex.org/C2983609787","wikidata":"https://www.wikidata.org/wiki/Q10534782","display_name":"Software implementation","level":3,"score":0.4328320622444153},{"id":"https://openalex.org/C149635348","wikidata":"https://www.wikidata.org/wiki/Q193040","display_name":"Embedded system","level":1,"score":0.429738312959671},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.2517184615135193},{"id":"https://openalex.org/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.12478438019752502}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3676536.3676667","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3676536.3676667","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3676536.3676667","source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 43rd IEEE/ACM International Conference on Computer-Aided Design","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.1145/3676536.3676667","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3676536.3676667","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3676536.3676667","source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 43rd IEEE/ACM International Conference on Computer-Aided Design","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G397489182","display_name":null,"funder_award_id":"2021YFA100-3602","funder_id":"https://openalex.org/F4320335777","funder_display_name":"National Key Research and Development Program of China"}],"funders":[{"id":"https://openalex.org/F4320321851","display_name":"Fudan University","ror":"https://ror.org/013q1eq08"},{"id":"https://openalex.org/F4320335777","display_name":"National Key Research and Development Program of China","ror":null}],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4409285558.pdf","grobid_xml":"https://content.openalex.org/works/W4409285558.grobid-xml"},"referenced_works_count":34,"referenced_works":["https://openalex.org/W398859631","https://openalex.org/W569478347","https://openalex.org/W2108598243","https://openalex.org/W2251610689","https://openalex.org/W2760788470","https://openalex.org/W2979455536","https://openalex.org/W3008408165","https://openalex.org/W3044749682","https://openalex.org/W3047848469","https://openalex.org/W3104151879","https://openalex.org/W3116489684","https://openalex.org/W3120586104","https://openalex.org/W3130240120","https://openalex.org/W3130641740","https://openalex.org/W3138516171","https://openalex.org/W3155487259","https://openalex.org/W3174350209","https://openalex.org/W4281708609","https://openalex.org/W4293025109","https://openalex.org/W4306295203","https://openalex.org/W4312772337","https://openalex.org/W4360831786","https://openalex.org/W4377864221","https://openalex.org/W4382765775","https://openalex.org/W4383749557","https://openalex.org/W4386083075","https://openalex.org/W4388214800","https://openalex.org/W4389162736","https://openalex.org/W4390872156","https://openalex.org/W6739901393","https://openalex.org/W6777017071","https://openalex.org/W6787300081","https://openalex.org/W6810353043","https://openalex.org/W6849805803"],"related_works":["https://openalex.org/W2111241003","https://openalex.org/W2355315220","https://openalex.org/W4200391368","https://openalex.org/W2210979487","https://openalex.org/W2316202402","https://openalex.org/W1967938402","https://openalex.org/W2386041993","https://openalex.org/W1608572506","https://openalex.org/W2883113558","https://openalex.org/W2556125783"],"abstract_inverted_index":{"Binarized":[0],"Transformer":[1,186],"(BiT)":[2],"networks":[3,38,71,228],"have":[4],"emerged":[5],"as":[6],"a":[7,29,45,65,87,97,113,221],"notable":[8],"advancement":[9],"in":[10],"artificial":[11],"intelligence,":[12],"reducing":[13,78],"computational":[14],"complexity":[15],"by":[16,151,206],"replacing":[17],"traditional":[18,156],"multiply-accumulate":[19],"operations":[20],"with":[21,64,143],"simpler":[22],"XNOR-popcount":[23],"operations.":[24,133],"This":[25,58],"paper":[26],"proposes":[27],"Edge-BiT,":[28],"framework":[30,224],"that":[31,138,229],"employs":[32],"software-hardware":[33,222],"co-design":[34,223],"to":[35,100,165,173,181],"optimize":[36],"BiT":[37,70,105,149,227],"inference":[39,79,102],"on":[40,81,96,220],"edge":[41],"FPGAs.":[42],"We":[43],"design":[44],"training":[46,140],"method":[47],"tailored":[48,129],"for":[49,130,185,225],"hardware":[50,82,109],"acceleration,":[51],"including":[52],"the":[53,108,146,210,217],"Exponential-Sparse":[54],"Data":[55],"Format":[56],"(ESDF).":[57],"format":[59],"is":[60,216],"not":[61],"only":[62],"compatible":[63],"broad":[66],"range":[67],"of":[68,148,163,199,212],"existing":[69,182],"but":[72],"also":[73],"enhances":[74],"their":[75],"accuracy":[76,147,233],"while":[77,192],"time":[80],"platforms.":[83],"Additionally,":[84],"we":[85],"develope":[86],"Compiler":[88],"and":[89,120,158,167,170,175,202,234],"an":[90,121],"Instruction":[91],"Set":[92],"Architecture":[93],"(ISA)":[94],"based":[95],"trigger-update":[98],"mechanism":[99],"support":[101],"across":[103],"various":[104,226],"networks.":[106],"On":[107],"side,":[110],"Edge-BiT":[111,153,188,215],"features":[112],"unique":[114],"Sparse":[115],"Matrix":[116],"XNOR":[117],"Unit":[118,125],"(SMXNOR)":[119],"optimized":[122],"Post":[123],"Process":[124],"(PPU),":[126],"both":[127,231],"specifically":[128],"efficient":[131],"binary":[132],"Our":[134],"empirical":[135],"evaluations":[136],"show":[137],"our":[139,213],"strategy,":[141],"coupled":[142],"ESDF,":[144],"increases":[145],"tasks":[150],"4.6--34.1%.":[152],"significantly":[154],"outperforms":[155],"CPU":[157],"GPU":[159],"solutions,":[160],"achieving":[161,196],"speedups":[162],"up":[164,172],"23.54\u00d7":[166],"9.34\u00d7,":[168],"respectively,":[169],"demonstrating":[171],"424.75\u00d7":[174],"121.91\u00d7":[176],"greater":[177],"energy":[178,204],"efficiency.":[179],"Compared":[180],"FPGA":[183],"accelerators":[184],"networks,":[187],"exhibits":[189],"superior":[190],"performance":[191],"utilizing":[193],"fewer":[194],"resources,":[195],"latency":[197],"reductions":[198],"1.16--2.78":[200],"\u00d7":[201],"improving":[203],"efficiency":[205],"1.34--7.43":[207],"\u00d7.":[208],"To":[209],"best":[211],"knowledge,":[214],"first":[218],"study":[219],"achieves":[230],"high":[232],"excellent":[235],"performance.":[236]},"counts_by_year":[{"year":2025,"cited_by_count":2}],"updated_date":"2025-11-06T03:46:38.306776","created_date":"2025-10-10T00:00:00"}