{"id":"https://openalex.org/W3115388607","doi":"https://doi.org/10.1109/tpds.2020.3047371","title":"Efficient Methods for Mapping Neural Machine Translator on FPGAs","display_name":"Efficient Methods for Mapping Neural Machine Translator on FPGAs","publication_year":2020,"publication_date":"2020-12-25","ids":{"openalex":"https://openalex.org/W3115388607","doi":"https://doi.org/10.1109/tpds.2020.3047371","mag":"3115388607"},"language":"en","primary_location":{"id":"doi:10.1109/tpds.2020.3047371","is_oa":true,"landing_page_url":"https://doi.org/10.1109/tpds.2020.3047371","pdf_url":"https://ieeexplore.ieee.org/ielx7/71/9357351/09309170.pdf","source":{"id":"https://openalex.org/S97130795","display_name":"IEEE Transactions on Parallel and Distributed Systems","issn_l":"1045-9219","issn":["1045-9219","1558-2183","2161-9883"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319808","host_organization_name":"Institute of Electrical and Electronics Engineers","host_organization_lineage":["https://openalex.org/P4310319808"],"host_organization_lineage_names":["Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Transactions on Parallel and Distributed Systems","raw_type":"journal-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"hybrid","oa_url":"https://ieeexplore.ieee.org/ielx7/71/9357351/09309170.pdf","any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5101584016","display_name":"Qin Li","orcid":"https://orcid.org/0000-0001-5220-9090"},"institutions":[{"id":"https://openalex.org/I157725225","display_name":"University of Illinois Urbana-Champaign","ror":"https://ror.org/047426m28","country_code":"US","type":"education","lineage":["https://openalex.org/I157725225"]}],"countries":["US"],"is_corresponding":true,"raw_author_name":"Qin Li","raw_affiliation_strings":["Department of Electrical and Computer Engineering, University of Illinois Urbana-Champaign, Champaign, IL, USA"],"affiliations":[{"raw_affiliation_string":"Department of Electrical and Computer Engineering, University of Illinois Urbana-Champaign, Champaign, IL, USA","institution_ids":["https://openalex.org/I157725225"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100330729","display_name":"Xiaofan Zhang","orcid":"https://orcid.org/0000-0001-5081-3972"},"institutions":[{"id":"https://openalex.org/I157725225","display_name":"University of Illinois Urbana-Champaign","ror":"https://ror.org/047426m28","country_code":"US","type":"education","lineage":["https://openalex.org/I157725225"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Xiaofan Zhang","raw_affiliation_strings":["Department of Electrical and Computer Engineering, University of Illinois Urbana-Champaign, Champaign, IL, USA"],"affiliations":[{"raw_affiliation_string":"Department of Electrical and Computer Engineering, University of Illinois Urbana-Champaign, Champaign, IL, USA","institution_ids":["https://openalex.org/I157725225"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5030156276","display_name":"Jinjun Xiong","orcid":"https://orcid.org/0000-0002-2620-4859"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jinjun Xiong","raw_affiliation_strings":["IBM T.J. Watson Research Center, NY, USA"],"affiliations":[{"raw_affiliation_string":"IBM T.J. Watson Research Center, NY, USA","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5040404999","display_name":"Wen\u2010mei Hwu","orcid":"https://orcid.org/0000-0003-2532-5349"},"institutions":[{"id":"https://openalex.org/I157725225","display_name":"University of Illinois Urbana-Champaign","ror":"https://ror.org/047426m28","country_code":"US","type":"education","lineage":["https://openalex.org/I157725225"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Wen-Mei Hwu","raw_affiliation_strings":["Department of Electrical and Computer Engineering, University of Illinois Urbana-Champaign, Champaign, IL, USA"],"affiliations":[{"raw_affiliation_string":"Department of Electrical and Computer Engineering, University of Illinois Urbana-Champaign, Champaign, IL, USA","institution_ids":["https://openalex.org/I157725225"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5056321228","display_name":"Deming Chen","orcid":"https://orcid.org/0000-0002-3016-0270"},"institutions":[{"id":"https://openalex.org/I157725225","display_name":"University of Illinois Urbana-Champaign","ror":"https://ror.org/047426m28","country_code":"US","type":"education","lineage":["https://openalex.org/I157725225"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Deming Chen","raw_affiliation_strings":["Department of Electrical and Computer Engineering, University of Illinois Urbana-Champaign, Champaign, IL, USA"],"affiliations":[{"raw_affiliation_string":"Department of Electrical and Computer Engineering, University of Illinois Urbana-Champaign, Champaign, IL, USA","institution_ids":["https://openalex.org/I157725225"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":5,"corresponding_author_ids":["https://openalex.org/A5101584016"],"corresponding_institution_ids":["https://openalex.org/I157725225"],"apc_list":null,"apc_paid":null,"fwci":1.6236,"has_fulltext":true,"cited_by_count":22,"citation_normalized_percentile":{"value":0.87672344,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":96,"max":99},"biblio":{"volume":"32","issue":"7","first_page":"1866","last_page":"1877"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10181","display_name":"Natural Language Processing Techniques","score":0.9975000023841858,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"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/T10181","display_name":"Natural Language Processing Techniques","score":0.9975000023841858,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"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":0.9973000288009644,"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/T10028","display_name":"Topic Modeling","score":0.9933000206947327,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"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/computer-science","display_name":"Computer science","score":0.8811138272285461},{"id":"https://openalex.org/keywords/field-programmable-gate-array","display_name":"Field-programmable gate array","score":0.7720961570739746},{"id":"https://openalex.org/keywords/machine-translation","display_name":"Machine translation","score":0.7539467811584473},{"id":"https://openalex.org/keywords/speedup","display_name":"Speedup","score":0.721377968788147},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.46350258588790894},{"id":"https://openalex.org/keywords/key","display_name":"Key (lock)","score":0.45514705777168274},{"id":"https://openalex.org/keywords/hardware-acceleration","display_name":"Hardware acceleration","score":0.4405447244644165},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.44052308797836304},{"id":"https://openalex.org/keywords/exploit","display_name":"Exploit","score":0.43508094549179077},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.4216534495353699},{"id":"https://openalex.org/keywords/computer-engineering","display_name":"Computer engineering","score":0.379299134016037},{"id":"https://openalex.org/keywords/computer-architecture","display_name":"Computer architecture","score":0.3648356795310974},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.3503812551498413},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.33007949590682983},{"id":"https://openalex.org/keywords/embedded-system","display_name":"Embedded system","score":0.3049551844596863}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8811138272285461},{"id":"https://openalex.org/C42935608","wikidata":"https://www.wikidata.org/wiki/Q190411","display_name":"Field-programmable gate array","level":2,"score":0.7720961570739746},{"id":"https://openalex.org/C203005215","wikidata":"https://www.wikidata.org/wiki/Q79798","display_name":"Machine translation","level":2,"score":0.7539467811584473},{"id":"https://openalex.org/C68339613","wikidata":"https://www.wikidata.org/wiki/Q1549489","display_name":"Speedup","level":2,"score":0.721377968788147},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.46350258588790894},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.45514705777168274},{"id":"https://openalex.org/C13164978","wikidata":"https://www.wikidata.org/wiki/Q600158","display_name":"Hardware acceleration","level":3,"score":0.4405447244644165},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.44052308797836304},{"id":"https://openalex.org/C165696696","wikidata":"https://www.wikidata.org/wiki/Q11287","display_name":"Exploit","level":2,"score":0.43508094549179077},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4216534495353699},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.379299134016037},{"id":"https://openalex.org/C118524514","wikidata":"https://www.wikidata.org/wiki/Q173212","display_name":"Computer architecture","level":1,"score":0.3648356795310974},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3503812551498413},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.33007949590682983},{"id":"https://openalex.org/C149635348","wikidata":"https://www.wikidata.org/wiki/Q193040","display_name":"Embedded system","level":1,"score":0.3049551844596863},{"id":"https://openalex.org/C38652104","wikidata":"https://www.wikidata.org/wiki/Q3510521","display_name":"Computer security","level":1,"score":0.0}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1109/tpds.2020.3047371","is_oa":true,"landing_page_url":"https://doi.org/10.1109/tpds.2020.3047371","pdf_url":"https://ieeexplore.ieee.org/ielx7/71/9357351/09309170.pdf","source":{"id":"https://openalex.org/S97130795","display_name":"IEEE Transactions on Parallel and Distributed Systems","issn_l":"1045-9219","issn":["1045-9219","1558-2183","2161-9883"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319808","host_organization_name":"Institute of Electrical and Electronics Engineers","host_organization_lineage":["https://openalex.org/P4310319808"],"host_organization_lineage_names":["Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Transactions on Parallel and Distributed Systems","raw_type":"journal-article"},{"id":"pmh:oai:www.ideals.illinois.edu:2142/108193","is_oa":false,"landing_page_url":"http://hdl.handle.net/2142/108193","pdf_url":null,"source":{"id":"https://openalex.org/S4377196349","display_name":"IDEALS (University of Illinois Urbana-Champaign)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I157725225","host_organization_name":"University of Illinois Urbana-Champaign","host_organization_lineage":["https://openalex.org/I157725225"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"","raw_type":"Thesis"}],"best_oa_location":{"id":"doi:10.1109/tpds.2020.3047371","is_oa":true,"landing_page_url":"https://doi.org/10.1109/tpds.2020.3047371","pdf_url":"https://ieeexplore.ieee.org/ielx7/71/9357351/09309170.pdf","source":{"id":"https://openalex.org/S97130795","display_name":"IEEE Transactions on Parallel and Distributed Systems","issn_l":"1045-9219","issn":["1045-9219","1558-2183","2161-9883"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319808","host_organization_name":"Institute of Electrical and Electronics Engineers","host_organization_lineage":["https://openalex.org/P4310319808"],"host_organization_lineage_names":["Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Transactions on Parallel and Distributed Systems","raw_type":"journal-article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W3115388607.pdf","grobid_xml":"https://content.openalex.org/works/W3115388607.grobid-xml"},"referenced_works_count":49,"referenced_works":["https://openalex.org/W22168010","https://openalex.org/W118713324","https://openalex.org/W179875071","https://openalex.org/W1753482797","https://openalex.org/W1924770834","https://openalex.org/W2018055497","https://openalex.org/W2064675550","https://openalex.org/W2094756095","https://openalex.org/W2130942839","https://openalex.org/W2133564696","https://openalex.org/W2137966900","https://openalex.org/W2276486856","https://openalex.org/W2294320930","https://openalex.org/W2525778437","https://openalex.org/W2552839021","https://openalex.org/W2585720638","https://openalex.org/W2588448445","https://openalex.org/W2727238169","https://openalex.org/W2762910930","https://openalex.org/W2788838111","https://openalex.org/W2887782013","https://openalex.org/W2889959382","https://openalex.org/W2899915146","https://openalex.org/W2915106038","https://openalex.org/W2962856739","https://openalex.org/W2962944188","https://openalex.org/W2963403868","https://openalex.org/W2964010167","https://openalex.org/W2964045208","https://openalex.org/W2964308564","https://openalex.org/W2974412363","https://openalex.org/W2981972129","https://openalex.org/W2998704965","https://openalex.org/W3003168380","https://openalex.org/W3037844041","https://openalex.org/W4285719527","https://openalex.org/W4385245566","https://openalex.org/W6600880057","https://openalex.org/W6604839928","https://openalex.org/W6607333740","https://openalex.org/W6637698695","https://openalex.org/W6640212811","https://openalex.org/W6679434410","https://openalex.org/W6679436768","https://openalex.org/W6680532216","https://openalex.org/W6739901393","https://openalex.org/W6767525790","https://openalex.org/W6769194895","https://openalex.org/W6898505805"],"related_works":["https://openalex.org/W2518118925","https://openalex.org/W2532502681","https://openalex.org/W3159273459","https://openalex.org/W4319952061","https://openalex.org/W4280636456","https://openalex.org/W4388913998","https://openalex.org/W4310584535","https://openalex.org/W4295935044","https://openalex.org/W4307927141","https://openalex.org/W3159906349"],"abstract_inverted_index":{"Neural":[0],"machine":[1],"translation":[2,61],"(NMT)":[3],"is":[4,211],"one":[5,24],"of":[6,20,40,96,111,119,170,199],"the":[7,17,65,69,94,109,132,149,160,185,212],"most":[8],"critical":[9],"applications":[10],"in":[11,23,165,173],"natural":[12],"language":[13,25],"processing":[14],"(NLP)":[15],"with":[16,64,116,195],"main":[18],"idea":[19],"converting":[21],"text":[22],"to":[26,80,92,98,122,162,184,223],"another":[27],"using":[28],"deep":[29],"neural":[30],"networks.":[31],"In":[32,87],"recent":[33],"year,":[34],"we":[35,90,191],"have":[36,73],"seen":[37],"continuous":[38],"development":[39],"NMT":[41,71,105,114,187,221],"by":[42],"integrating":[43],"more":[44,76],"emerging":[45],"technologies,":[46],"such":[47],"as":[48,180],"bidirectional":[49,150],"gated":[50],"recurrent":[51],"units":[52],"(GRU),":[53],"attention":[54,153],"mechanisms,":[55],"and":[56,78,130,155,168,177],"beam-search":[57],"algorithms,":[58],"for":[59,84,103,146],"improved":[60],"quality.":[62],"However,":[63],"increasing":[66],"problem":[67],"size,":[68],"real-life":[70,104,219],"models":[72],"become":[74],"much":[75],"complicated":[77],"difficult":[79],"implement":[81],"on":[82,189,207,226],"hardware":[83],"acceleration":[85],"opportunities.":[86],"this":[88,210],"article,":[89],"aim":[91],"exploit":[93],"capability":[95],"FPGAs":[97],"deliver":[99],"highly":[100,124],"efficient":[101],"implementations":[102],"applications.":[106],"We":[107,158],"map":[108],"inference":[110],"a":[112,123,218],"large-scale":[113],"model":[115,140,161,222],"total":[117],"computation":[118],"172":[120],"GFLOP":[121],"optimized":[125],"high-level":[126],"synthesis":[127],"(HLS)":[128],"IP":[129,133],"integrate":[131],"into":[134],"Xilinx":[135],"VCU118":[136],"FPGA":[137,225],"platform.":[138],"The":[139],"has":[141],"widely":[142],"used":[143],"key":[144],"features":[145],"NMTs,":[147],"including":[148],"GRU":[151],"layer,":[152],"mechanism,":[154],"beam":[156],"search.":[157],"quantize":[159],"mixed-precision":[163],"representation":[164],"which":[166],"parameters":[167],"portions":[169],"calculations":[171],"are":[172],"16-bit":[174],"half":[175],"precision,":[176],"others":[178],"remain":[179],"32-bit":[181],"floating-point.":[182],"Compared":[183],"float":[186],"implementation":[188],"FPGA,":[190],"achieve":[192],"13.1\u00d7":[193],"speedup":[194],"an":[196,224],"end-to-end":[197,220],"performance":[198],"22.0":[200],"GFLOPS":[201],"without":[202],"any":[203],"accuracy":[204],"degradation.":[205],"Based":[206],"our":[208],"knowledge,":[209],"first":[213],"work":[214],"that":[215],"successfully":[216],"implements":[217],"board.":[227]},"counts_by_year":[{"year":2026,"cited_by_count":1},{"year":2025,"cited_by_count":4},{"year":2024,"cited_by_count":5},{"year":2023,"cited_by_count":4},{"year":2022,"cited_by_count":3},{"year":2021,"cited_by_count":5}],"updated_date":"2026-04-04T16:13:02.066488","created_date":"2025-10-10T00:00:00"}