{"id":"https://openalex.org/W3198028914","doi":"https://doi.org/10.1162/coli_a_00421","title":"Sequence-Level Training for Non-Autoregressive Neural Machine Translation","display_name":"Sequence-Level Training for Non-Autoregressive Neural Machine Translation","publication_year":2021,"publication_date":"2021-09-02","ids":{"openalex":"https://openalex.org/W3198028914","doi":"https://doi.org/10.1162/coli_a_00421","mag":"3198028914"},"language":"en","primary_location":{"id":"doi:10.1162/coli_a_00421","is_oa":true,"landing_page_url":"https://doi.org/10.1162/coli_a_00421","pdf_url":"https://direct.mit.edu/coli/article-pdf/47/4/891/1979393/coli_a_00421.pdf","source":{"id":"https://openalex.org/S155526855","display_name":"Computational Linguistics","issn_l":"0891-2017","issn":["0891-2017","1530-9312"],"is_oa":true,"is_in_doaj":true,"is_core":true,"host_organization":"https://openalex.org/P4310320244","host_organization_name":"Association for Computational Linguistics","host_organization_lineage":["https://openalex.org/P4310320244"],"host_organization_lineage_names":["Association for Computational Linguistics"],"type":"journal"},"license":"cc-by-nc-nd","license_id":"https://openalex.org/licenses/cc-by-nc-nd","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Computational Linguistics","raw_type":"journal-article"},"type":"article","indexed_in":["crossref","doaj"],"open_access":{"is_oa":true,"oa_status":"diamond","oa_url":"https://direct.mit.edu/coli/article-pdf/47/4/891/1979393/coli_a_00421.pdf","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5056500539","display_name":"Chenze Shao","orcid":"https://orcid.org/0000-0001-7724-0684"},"institutions":[{"id":"https://openalex.org/I19820366","display_name":"Chinese Academy of Sciences","ror":"https://ror.org/034t30j35","country_code":"CN","type":"government","lineage":["https://openalex.org/I19820366"]},{"id":"https://openalex.org/I4210090176","display_name":"Institute of Computing Technology","ror":"https://ror.org/0090r4d87","country_code":"CN","type":"facility","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210090176"]}],"countries":["CN"],"is_corresponding":true,"raw_author_name":"Chenze Shao","raw_affiliation_strings":["Key Laboratory of Intelligent Information Processing, Institute of Computing Technology, Chinese Academy of Sciences. shaochenze18z@ict.ac.cn"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Key Laboratory of Intelligent Information Processing, Institute of Computing Technology, Chinese Academy of Sciences. shaochenze18z@ict.ac.cn","institution_ids":["https://openalex.org/I4210090176","https://openalex.org/I19820366"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5000232528","display_name":"Yang Feng","orcid":"https://orcid.org/0000-0002-1898-3784"},"institutions":[{"id":"https://openalex.org/I19820366","display_name":"Chinese Academy of Sciences","ror":"https://ror.org/034t30j35","country_code":"CN","type":"government","lineage":["https://openalex.org/I19820366"]},{"id":"https://openalex.org/I4210090176","display_name":"Institute of Computing Technology","ror":"https://ror.org/0090r4d87","country_code":"CN","type":"facility","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210090176"]}],"countries":["CN"],"is_corresponding":true,"raw_author_name":"Yang Feng","raw_affiliation_strings":["Key Laboratory of Intelligent Information Processing, Institute of Computing Technology, Chinese Academy of Sciences. fengyang@ict.ac.cn"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Key Laboratory of Intelligent Information Processing, Institute of Computing Technology, Chinese Academy of Sciences. fengyang@ict.ac.cn","institution_ids":["https://openalex.org/I4210090176","https://openalex.org/I19820366"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5076799733","display_name":"Jinchao Zhang","orcid":"https://orcid.org/0000-0002-5279-0468"},"institutions":[{"id":"https://openalex.org/I2250653659","display_name":"Tencent (China)","ror":"https://ror.org/00hhjss72","country_code":"CN","type":"company","lineage":["https://openalex.org/I2250653659"]}],"countries":["CN"],"is_corresponding":true,"raw_author_name":"Jinchao Zhang","raw_affiliation_strings":["Pattern Recognition Center, WeChat AI, Tencent Inc. dayerzhang@tencent.com"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Pattern Recognition Center, WeChat AI, Tencent Inc. dayerzhang@tencent.com","institution_ids":["https://openalex.org/I2250653659"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5024849044","display_name":"Fandong Meng","orcid":"https://orcid.org/0000-0002-8158-2377"},"institutions":[{"id":"https://openalex.org/I2250653659","display_name":"Tencent (China)","ror":"https://ror.org/00hhjss72","country_code":"CN","type":"company","lineage":["https://openalex.org/I2250653659"]}],"countries":["CN"],"is_corresponding":true,"raw_author_name":"Fandong Meng","raw_affiliation_strings":["Pattern Recognition Center, WeChat AI, Tencent Inc. fandongmeng@tencent.com"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Pattern Recognition Center, WeChat AI, Tencent Inc. fandongmeng@tencent.com","institution_ids":["https://openalex.org/I2250653659"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5024102664","display_name":"Jie Zhou","orcid":"https://orcid.org/0000-0002-3371-6780"},"institutions":[{"id":"https://openalex.org/I2250653659","display_name":"Tencent (China)","ror":"https://ror.org/00hhjss72","country_code":"CN","type":"company","lineage":["https://openalex.org/I2250653659"]}],"countries":["CN"],"is_corresponding":true,"raw_author_name":"Jie Zhou","raw_affiliation_strings":["Pattern Recognition Center, WeChat AI, Tencent Inc. withtomzhou@tencent.com"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Pattern Recognition Center, WeChat AI, Tencent Inc. withtomzhou@tencent.com","institution_ids":["https://openalex.org/I2250653659"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":5,"corresponding_author_ids":["https://openalex.org/A5000232528","https://openalex.org/A5024102664","https://openalex.org/A5024849044","https://openalex.org/A5056500539","https://openalex.org/A5076799733"],"corresponding_institution_ids":["https://openalex.org/I19820366","https://openalex.org/I2250653659","https://openalex.org/I4210090176"],"apc_list":null,"apc_paid":null,"fwci":2.2386,"has_fulltext":false,"cited_by_count":20,"citation_normalized_percentile":{"value":0.90075439,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":95,"max":99},"biblio":{"volume":"47","issue":"4","first_page":"891","last_page":"925"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10181","display_name":"Natural Language Processing Techniques","score":1.0,"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":1.0,"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/T10028","display_name":"Topic Modeling","score":0.9995999932289124,"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/T11714","display_name":"Multimodal Machine Learning Applications","score":0.987500011920929,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.8658697009086609},{"id":"https://openalex.org/keywords/machine-translation","display_name":"Machine translation","score":0.8294475078582764},{"id":"https://openalex.org/keywords/autoregressive-model","display_name":"Autoregressive model","score":0.7029123306274414},{"id":"https://openalex.org/keywords/nat","display_name":"Nat","score":0.6899493932723999},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.5722405314445496},{"id":"https://openalex.org/keywords/translation","display_name":"Translation (biology)","score":0.4846106469631195},{"id":"https://openalex.org/keywords/word","display_name":"Word (group theory)","score":0.44969895482063293},{"id":"https://openalex.org/keywords/speedup","display_name":"Speedup","score":0.4408002197742462},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.4265438914299011},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.4167991876602173},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.37139421701431274},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.35414379835128784},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.09133997559547424},{"id":"https://openalex.org/keywords/statistics","display_name":"Statistics","score":0.08896377682685852}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8658697009086609},{"id":"https://openalex.org/C203005215","wikidata":"https://www.wikidata.org/wiki/Q79798","display_name":"Machine translation","level":2,"score":0.8294475078582764},{"id":"https://openalex.org/C159877910","wikidata":"https://www.wikidata.org/wiki/Q2202883","display_name":"Autoregressive model","level":2,"score":0.7029123306274414},{"id":"https://openalex.org/C182516595","wikidata":"https://www.wikidata.org/wiki/Q376660","display_name":"Nat","level":2,"score":0.6899493932723999},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5722405314445496},{"id":"https://openalex.org/C149364088","wikidata":"https://www.wikidata.org/wiki/Q185917","display_name":"Translation (biology)","level":4,"score":0.4846106469631195},{"id":"https://openalex.org/C90805587","wikidata":"https://www.wikidata.org/wiki/Q10944557","display_name":"Word (group theory)","level":2,"score":0.44969895482063293},{"id":"https://openalex.org/C68339613","wikidata":"https://www.wikidata.org/wiki/Q1549489","display_name":"Speedup","level":2,"score":0.4408002197742462},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.4265438914299011},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.4167991876602173},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.37139421701431274},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.35414379835128784},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.09133997559547424},{"id":"https://openalex.org/C105795698","wikidata":"https://www.wikidata.org/wiki/Q12483","display_name":"Statistics","level":1,"score":0.08896377682685852},{"id":"https://openalex.org/C31258907","wikidata":"https://www.wikidata.org/wiki/Q1301371","display_name":"Computer network","level":1,"score":0.0},{"id":"https://openalex.org/C41895202","wikidata":"https://www.wikidata.org/wiki/Q8162","display_name":"Linguistics","level":1,"score":0.0},{"id":"https://openalex.org/C185592680","wikidata":"https://www.wikidata.org/wiki/Q2329","display_name":"Chemistry","level":0,"score":0.0},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.0},{"id":"https://openalex.org/C104317684","wikidata":"https://www.wikidata.org/wiki/Q7187","display_name":"Gene","level":2,"score":0.0},{"id":"https://openalex.org/C105580179","wikidata":"https://www.wikidata.org/wiki/Q188928","display_name":"Messenger RNA","level":3,"score":0.0},{"id":"https://openalex.org/C55493867","wikidata":"https://www.wikidata.org/wiki/Q7094","display_name":"Biochemistry","level":1,"score":0.0},{"id":"https://openalex.org/C138885662","wikidata":"https://www.wikidata.org/wiki/Q5891","display_name":"Philosophy","level":0,"score":0.0}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1162/coli_a_00421","is_oa":true,"landing_page_url":"https://doi.org/10.1162/coli_a_00421","pdf_url":"https://direct.mit.edu/coli/article-pdf/47/4/891/1979393/coli_a_00421.pdf","source":{"id":"https://openalex.org/S155526855","display_name":"Computational Linguistics","issn_l":"0891-2017","issn":["0891-2017","1530-9312"],"is_oa":true,"is_in_doaj":true,"is_core":true,"host_organization":"https://openalex.org/P4310320244","host_organization_name":"Association for Computational Linguistics","host_organization_lineage":["https://openalex.org/P4310320244"],"host_organization_lineage_names":["Association for Computational Linguistics"],"type":"journal"},"license":"cc-by-nc-nd","license_id":"https://openalex.org/licenses/cc-by-nc-nd","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Computational Linguistics","raw_type":"journal-article"},{"id":"pmh:oai:doaj.org/article:2cf968bfda0f483ab16c66e22add3fcf","is_oa":true,"landing_page_url":"https://doaj.org/article/2cf968bfda0f483ab16c66e22add3fcf","pdf_url":null,"source":{"id":"https://openalex.org/S4306401280","display_name":"DOAJ (DOAJ: Directory of Open Access Journals)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by-sa","license_id":"https://openalex.org/licenses/cc-by-sa","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"Computational Linguistics, Vol 47, Iss 4, Pp 891-925 (2021)","raw_type":"article"}],"best_oa_location":{"id":"doi:10.1162/coli_a_00421","is_oa":true,"landing_page_url":"https://doi.org/10.1162/coli_a_00421","pdf_url":"https://direct.mit.edu/coli/article-pdf/47/4/891/1979393/coli_a_00421.pdf","source":{"id":"https://openalex.org/S155526855","display_name":"Computational Linguistics","issn_l":"0891-2017","issn":["0891-2017","1530-9312"],"is_oa":true,"is_in_doaj":true,"is_core":true,"host_organization":"https://openalex.org/P4310320244","host_organization_name":"Association for Computational Linguistics","host_organization_lineage":["https://openalex.org/P4310320244"],"host_organization_lineage_names":["Association for Computational Linguistics"],"type":"journal"},"license":"cc-by-nc-nd","license_id":"https://openalex.org/licenses/cc-by-nc-nd","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Computational Linguistics","raw_type":"journal-article"},"sustainable_development_goals":[{"score":0.5099999904632568,"id":"https://metadata.un.org/sdg/4","display_name":"Quality Education"}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":82,"referenced_works":["https://openalex.org/W1522301498","https://openalex.org/W1821462560","https://openalex.org/W2016589492","https://openalex.org/W2101105183","https://openalex.org/W2119717200","https://openalex.org/W2149684865","https://openalex.org/W2157331557","https://openalex.org/W2166706824","https://openalex.org/W2499811908","https://openalex.org/W2525778437","https://openalex.org/W2542835211","https://openalex.org/W2559655401","https://openalex.org/W2601324753","https://openalex.org/W2747329762","https://openalex.org/W2887428522","https://openalex.org/W2888442053","https://openalex.org/W2890220768","https://openalex.org/W2890501761","https://openalex.org/W2892213699","https://openalex.org/W2912083425","https://openalex.org/W2912269676","https://openalex.org/W2913838200","https://openalex.org/W2914304175","https://openalex.org/W2914782308","https://openalex.org/W2915480215","https://openalex.org/W2948197522","https://openalex.org/W2952356761","https://openalex.org/W2962784628","https://openalex.org/W2962915948","https://openalex.org/W2962969034","https://openalex.org/W2963246629","https://openalex.org/W2963434219","https://openalex.org/W2963463964","https://openalex.org/W2963536265","https://openalex.org/W2963736842","https://openalex.org/W2964190861","https://openalex.org/W2964345285","https://openalex.org/W2970832665","https://openalex.org/W2976965654","https://openalex.org/W2988975212","https://openalex.org/W2990004017","https://openalex.org/W2996987694","https://openalex.org/W2997284037","https://openalex.org/W3034216012","https://openalex.org/W3034425996","https://openalex.org/W3034579764","https://openalex.org/W3100753857","https://openalex.org/W3106104873","https://openalex.org/W3175092301","https://openalex.org/W4212774754","https://openalex.org/W6621543089","https://openalex.org/W6631190155","https://openalex.org/W6638088447","https://openalex.org/W6638523607","https://openalex.org/W6679434410","https://openalex.org/W6679436768","https://openalex.org/W6682631176","https://openalex.org/W6682888085","https://openalex.org/W6683204974","https://openalex.org/W6685322675","https://openalex.org/W6693958471","https://openalex.org/W6722524744","https://openalex.org/W6725107885","https://openalex.org/W6727690538","https://openalex.org/W6727862155","https://openalex.org/W6729383884","https://openalex.org/W6732865468","https://openalex.org/W6736588030","https://openalex.org/W6737778391","https://openalex.org/W6739901393","https://openalex.org/W6746208923","https://openalex.org/W6748381668","https://openalex.org/W6755207826","https://openalex.org/W6762471145","https://openalex.org/W6767128473","https://openalex.org/W6769692749","https://openalex.org/W6769921281","https://openalex.org/W6769953141","https://openalex.org/W6770755217","https://openalex.org/W6775291257","https://openalex.org/W6780062628","https://openalex.org/W6780230508"],"related_works":["https://openalex.org/W2383430300","https://openalex.org/W2355747264","https://openalex.org/W2058965144","https://openalex.org/W2164382479","https://openalex.org/W98480971","https://openalex.org/W2150291671","https://openalex.org/W2027972911","https://openalex.org/W2146343568","https://openalex.org/W2728761353","https://openalex.org/W2883671469"],"abstract_inverted_index":{"Abstract":[0],"In":[1,90],"recent":[2],"years,":[3],"Neural":[4,40],"Machine":[5,41],"Translation":[6,42],"(NMT)":[7],"has":[8],"achieved":[9],"notable":[10],"results":[11],"in":[12],"various":[13],"translation":[14,29,117,217,238],"tasks.":[15,239],"However,":[16],"the":[17,23,32,45,64,69,77,87,105,115,143,148,167,172,176,208],"word-by-word":[18],"generation":[19],"manner":[20],"determined":[21],"by":[22,53,146],"autoregressive":[24,46],"mechanism":[25,47],"leads":[26],"to":[27,86,99,125,165,201,206],"high":[28],"latency":[30],"of":[31,79,150],"NMT":[33],"and":[34,48,58,111,175,185,232],"restricts":[35],"its":[36],"low-latency":[37],"applications.":[38],"Non-Autoregressive":[39],"(NAT)":[43],"removes":[44],"achieves":[49,233],"significant":[50],"decoding":[51],"speedup":[52],"generating":[54],"target":[55],"words":[56],"independently":[57],"simultaneously.":[59],"Nevertheless,":[60],"NAT":[61,80,101,106,123,161,209,230],"still":[62],"takes":[63],"word-level":[65],"cross-entropy":[66],"loss":[67],"as":[68,108],"training":[70,97,122,158,181,199],"objective,":[71],"which":[72,103,141,163,223],"is":[73,183,243],"not":[74],"optimal":[75],"because":[76],"output":[78,174],"cannot":[81],"be":[82,187],"properly":[83],"evaluated":[84],"due":[85],"multimodality":[88],"problem.":[89],"this":[91],"article,":[92],"we":[93,120,154,195],"propose":[94,121],"using":[95],"sequence-level":[96,127],"objectives":[98],"train":[100,207],"models,":[102,162],"evaluate":[104],"outputs":[107],"a":[109,156,197],"whole":[110],"correlates":[112],"well":[113],"with":[114],"real":[116],"quality.":[118],"First,":[119],"models":[124],"optimize":[126],"evaluation":[128],"metrics":[129],"(e.g.,":[130],"BLEU)":[131],"based":[132],"on":[133,215,236],"several":[134],"novel":[135,157],"reinforcement":[136],"algorithms":[137],"customized":[138],"for":[139,160],"NAT,":[140],"outperform":[142],"conventional":[144],"method":[145],"reducing":[147],"variance":[149],"gradient":[151],"estimation.":[152],"Second,":[153],"introduce":[155],"objective":[159,182],"aims":[164],"minimize":[166],"Bag-of-N-grams":[168],"(BoN)":[169],"difference":[170],"between":[171],"model":[173],"reference":[177],"sentence.":[178],"The":[179,240],"BoN":[180],"differentiable":[184],"can":[186],"calculated":[188],"efficiently":[189],"without":[190],"doing":[191],"any":[192],"approximations.":[193],"Finally,":[194],"apply":[196],"three-stage":[198],"strategy":[200],"combine":[202],"these":[203],"two":[204],"methods":[205],"model.":[210],"We":[211],"validate":[212],"our":[213,226],"approach":[214,227],"four":[216],"tasks":[218],"(WMT14":[219],"En\u2194De,":[220],"WMT16":[221],"En\u2194Ro),":[222],"shows":[224],"that":[225],"largely":[228],"outperforms":[229],"baselines":[231],"remarkable":[234],"performance":[235],"all":[237],"source":[241],"code":[242],"available":[244],"at":[245],"https://github.com/ictnlp/Seq-NAT.":[246]},"counts_by_year":[{"year":2026,"cited_by_count":2},{"year":2025,"cited_by_count":2},{"year":2024,"cited_by_count":4},{"year":2023,"cited_by_count":6},{"year":2022,"cited_by_count":6}],"updated_date":"2026-05-21T06:26:12.895304","created_date":"2025-10-10T00:00:00"}