{"id":"https://openalex.org/W3162226363","doi":"https://doi.org/10.1109/taslp.2021.3076863","title":"Adaptive Adapters: An Efficient Way to Incorporate BERT Into Neural Machine Translation","display_name":"Adaptive Adapters: An Efficient Way to Incorporate BERT Into Neural Machine Translation","publication_year":2021,"publication_date":"2021-01-01","ids":{"openalex":"https://openalex.org/W3162226363","doi":"https://doi.org/10.1109/taslp.2021.3076863","mag":"3162226363"},"language":"en","primary_location":{"id":"doi:10.1109/taslp.2021.3076863","is_oa":false,"landing_page_url":"https://doi.org/10.1109/taslp.2021.3076863","pdf_url":null,"source":{"id":"https://openalex.org/S4210169297","display_name":"IEEE/ACM Transactions on Audio Speech and Language Processing","issn_l":"2329-9290","issn":["2329-9290","2329-9304"],"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":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE/ACM Transactions on Audio, Speech, and Language Processing","raw_type":"journal-article"},"type":"article","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/A5055122985","display_name":"Junliang Guo","orcid":"https://orcid.org/0000-0001-8360-5483"},"institutions":[{"id":"https://openalex.org/I126520041","display_name":"University of Science and Technology of China","ror":"https://ror.org/04c4dkn09","country_code":"CN","type":"education","lineage":["https://openalex.org/I126520041","https://openalex.org/I19820366"]}],"countries":["CN"],"is_corresponding":true,"raw_author_name":"Junliang Guo","raw_affiliation_strings":["Anhui Province Key Laboratory of Big Data Analysis and Application, School of Computer Science and Technology, University of Science and Technology of China, Hefei, China"],"affiliations":[{"raw_affiliation_string":"Anhui Province Key Laboratory of Big Data Analysis and Application, School of Computer Science and Technology, University of Science and Technology of China, Hefei, China","institution_ids":["https://openalex.org/I126520041"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5028604823","display_name":"Zhirui Zhang","orcid":"https://orcid.org/0000-0003-1385-3742"},"institutions":[{"id":"https://openalex.org/I45928872","display_name":"Alibaba Group (China)","ror":"https://ror.org/00k642b80","country_code":"CN","type":"company","lineage":["https://openalex.org/I45928872"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zhirui Zhang","raw_affiliation_strings":["Alibaba Damo Academy, Hangzhou, China"],"affiliations":[{"raw_affiliation_string":"Alibaba Damo Academy, Hangzhou, China","institution_ids":["https://openalex.org/I45928872"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5009732907","display_name":"Linli Xu","orcid":"https://orcid.org/0000-0003-0227-3793"},"institutions":[{"id":"https://openalex.org/I126520041","display_name":"University of Science and Technology of China","ror":"https://ror.org/04c4dkn09","country_code":"CN","type":"education","lineage":["https://openalex.org/I126520041","https://openalex.org/I19820366"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Linli Xu","raw_affiliation_strings":["Anhui Province Key Laboratory of Big Data Analysis and Application, School of Computer Science and Technology, University of Science, and Technology of China, Hefei, China","IFLYTEK Co., Ltd, Hefei Anhui, China"],"affiliations":[{"raw_affiliation_string":"Anhui Province Key Laboratory of Big Data Analysis and Application, School of Computer Science and Technology, University of Science, and Technology of China, Hefei, China","institution_ids":["https://openalex.org/I126520041"]},{"raw_affiliation_string":"IFLYTEK Co., Ltd, Hefei Anhui, China","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5019715118","display_name":"Boxing Chen","orcid":"https://orcid.org/0000-0002-3170-4858"},"institutions":[{"id":"https://openalex.org/I45928872","display_name":"Alibaba Group (China)","ror":"https://ror.org/00k642b80","country_code":"CN","type":"company","lineage":["https://openalex.org/I45928872"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Boxing Chen","raw_affiliation_strings":["Alibaba Damo Academy, Hangzhou, China"],"affiliations":[{"raw_affiliation_string":"Alibaba Damo Academy, Hangzhou, China","institution_ids":["https://openalex.org/I45928872"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5048237545","display_name":"Enhong Chen","orcid":"https://orcid.org/0000-0002-4835-4102"},"institutions":[{"id":"https://openalex.org/I126520041","display_name":"University of Science and Technology of China","ror":"https://ror.org/04c4dkn09","country_code":"CN","type":"education","lineage":["https://openalex.org/I126520041","https://openalex.org/I19820366"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Enhong Chen","raw_affiliation_strings":["Anhui Province Key Laboratory of Big Data Analysis and Application, School of Computer Science and Technology, University of Science and Technology of China, Hefei, China"],"affiliations":[{"raw_affiliation_string":"Anhui Province Key Laboratory of Big Data Analysis and Application, School of Computer Science and Technology, University of Science and Technology of China, Hefei, China","institution_ids":["https://openalex.org/I126520041"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":5,"corresponding_author_ids":["https://openalex.org/A5055122985"],"corresponding_institution_ids":["https://openalex.org/I126520041"],"apc_list":null,"apc_paid":null,"fwci":2.6597,"has_fulltext":false,"cited_by_count":30,"citation_normalized_percentile":{"value":0.91476061,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":95,"max":99},"biblio":{"volume":"29","issue":null,"first_page":"1740","last_page":"1751"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.9998000264167786,"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/T10028","display_name":"Topic Modeling","score":0.9998000264167786,"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/T10181","display_name":"Natural Language Processing Techniques","score":0.9998000264167786,"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/T10201","display_name":"Speech Recognition and Synthesis","score":0.9868999719619751,"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":0.8657115697860718},{"id":"https://openalex.org/keywords/machine-translation","display_name":"Machine translation","score":0.8422072529792786},{"id":"https://openalex.org/keywords/language-model","display_name":"Language model","score":0.7612829208374023},{"id":"https://openalex.org/keywords/adapter","display_name":"Adapter (computing)","score":0.7456842064857483},{"id":"https://openalex.org/keywords/autoregressive-model","display_name":"Autoregressive model","score":0.6670606136322021},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.5996346473693848},{"id":"https://openalex.org/keywords/decoding-methods","display_name":"Decoding methods","score":0.5904406905174255},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.5236831903457642},{"id":"https://openalex.org/keywords/speedup","display_name":"Speedup","score":0.45300590991973877},{"id":"https://openalex.org/keywords/latency","display_name":"Latency (audio)","score":0.4351273775100708},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.35872945189476013},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.34272706508636475},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.18501025438308716},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.153077632188797}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8657115697860718},{"id":"https://openalex.org/C203005215","wikidata":"https://www.wikidata.org/wiki/Q79798","display_name":"Machine translation","level":2,"score":0.8422072529792786},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.7612829208374023},{"id":"https://openalex.org/C177284502","wikidata":"https://www.wikidata.org/wiki/Q1005390","display_name":"Adapter (computing)","level":2,"score":0.7456842064857483},{"id":"https://openalex.org/C159877910","wikidata":"https://www.wikidata.org/wiki/Q2202883","display_name":"Autoregressive model","level":2,"score":0.6670606136322021},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.5996346473693848},{"id":"https://openalex.org/C57273362","wikidata":"https://www.wikidata.org/wiki/Q576722","display_name":"Decoding methods","level":2,"score":0.5904406905174255},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5236831903457642},{"id":"https://openalex.org/C68339613","wikidata":"https://www.wikidata.org/wiki/Q1549489","display_name":"Speedup","level":2,"score":0.45300590991973877},{"id":"https://openalex.org/C82876162","wikidata":"https://www.wikidata.org/wiki/Q17096504","display_name":"Latency (audio)","level":2,"score":0.4351273775100708},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.35872945189476013},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.34272706508636475},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.18501025438308716},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.153077632188797},{"id":"https://openalex.org/C76155785","wikidata":"https://www.wikidata.org/wiki/Q418","display_name":"Telecommunications","level":1,"score":0.0},{"id":"https://openalex.org/C149782125","wikidata":"https://www.wikidata.org/wiki/Q160039","display_name":"Econometrics","level":1,"score":0.0},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.0},{"id":"https://openalex.org/C162324750","wikidata":"https://www.wikidata.org/wiki/Q8134","display_name":"Economics","level":0,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/taslp.2021.3076863","is_oa":false,"landing_page_url":"https://doi.org/10.1109/taslp.2021.3076863","pdf_url":null,"source":{"id":"https://openalex.org/S4210169297","display_name":"IEEE/ACM Transactions on Audio Speech and Language Processing","issn_l":"2329-9290","issn":["2329-9290","2329-9304"],"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":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE/ACM Transactions on Audio, Speech, and Language Processing","raw_type":"journal-article"}],"best_oa_location":null,"sustainable_development_goals":[{"score":0.7900000214576721,"display_name":"Quality Education","id":"https://metadata.un.org/sdg/4"}],"awards":[{"id":"https://openalex.org/G6481167376","display_name":null,"funder_award_id":"2008085J31","funder_id":"https://openalex.org/F4320334897","funder_display_name":"Natural Science Foundation of Anhui Province"},{"id":"https://openalex.org/G6859851492","display_name":null,"funder_award_id":"U20A20229","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"}],"funders":[{"id":"https://openalex.org/F4320321001","display_name":"National Natural Science Foundation of China","ror":"https://ror.org/01h0zpd94"},{"id":"https://openalex.org/F4320334897","display_name":"Natural Science Foundation of Anhui Province","ror":null}],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":116,"referenced_works":["https://openalex.org/W1682403713","https://openalex.org/W1821462560","https://openalex.org/W1959608418","https://openalex.org/W2101105183","https://openalex.org/W2114766824","https://openalex.org/W2133564696","https://openalex.org/W2194775991","https://openalex.org/W2225156818","https://openalex.org/W2547875792","https://openalex.org/W2767206889","https://openalex.org/W2889326796","https://openalex.org/W2896409484","https://openalex.org/W2896457183","https://openalex.org/W2914120296","https://openalex.org/W2920538220","https://openalex.org/W2923014074","https://openalex.org/W2924902521","https://openalex.org/W2944815030","https://openalex.org/W2945260553","https://openalex.org/W2946375144","https://openalex.org/W2946794439","https://openalex.org/W2949644922","https://openalex.org/W2951569836","https://openalex.org/W2952468927","https://openalex.org/W2962739339","https://openalex.org/W2962965870","https://openalex.org/W2962969034","https://openalex.org/W2963000224","https://openalex.org/W2963109507","https://openalex.org/W2963211188","https://openalex.org/W2963310665","https://openalex.org/W2963341956","https://openalex.org/W2963363373","https://openalex.org/W2963403868","https://openalex.org/W2963736842","https://openalex.org/W2963748441","https://openalex.org/W2963929190","https://openalex.org/W2963993763","https://openalex.org/W2964303773","https://openalex.org/W2964308564","https://openalex.org/W2965373594","https://openalex.org/W2970049541","https://openalex.org/W2970597249","https://openalex.org/W2970925270","https://openalex.org/W2971167892","https://openalex.org/W2971274815","https://openalex.org/W2972451902","https://openalex.org/W2975185270","https://openalex.org/W2975381464","https://openalex.org/W2978017171","https://openalex.org/W2981648103","https://openalex.org/W2986367395","https://openalex.org/W2988975212","https://openalex.org/W2990389671","https://openalex.org/W2994928925","https://openalex.org/W2995998574","https://openalex.org/W2996159613","https://openalex.org/W2996854111","https://openalex.org/W2996987694","https://openalex.org/W2997518171","https://openalex.org/W2997763445","https://openalex.org/W3005441132","https://openalex.org/W3006381853","https://openalex.org/W3015609966","https://openalex.org/W3030163527","https://openalex.org/W3034999214","https://openalex.org/W3035289598","https://openalex.org/W3035317912","https://openalex.org/W3036120435","https://openalex.org/W3038012435","https://openalex.org/W3089072946","https://openalex.org/W3093345276","https://openalex.org/W3100311862","https://openalex.org/W3101380508","https://openalex.org/W3101731278","https://openalex.org/W3105281812","https://openalex.org/W3105966348","https://openalex.org/W3175604467","https://openalex.org/W4292779060","https://openalex.org/W4385245566","https://openalex.org/W6638523607","https://openalex.org/W6640963894","https://openalex.org/W6677103964","https://openalex.org/W6679434410","https://openalex.org/W6718053083","https://openalex.org/W6725543821","https://openalex.org/W6726275242","https://openalex.org/W6729448088","https://openalex.org/W6738045163","https://openalex.org/W6739901393","https://openalex.org/W6746208923","https://openalex.org/W6748979873","https://openalex.org/W6755207826","https://openalex.org/W6755843862","https://openalex.org/W6759455113","https://openalex.org/W6759579507","https://openalex.org/W6760017253","https://openalex.org/W6760732026","https://openalex.org/W6762122294","https://openalex.org/W6762471145","https://openalex.org/W6762521896","https://openalex.org/W6766673545","https://openalex.org/W6768080748","https://openalex.org/W6768086466","https://openalex.org/W6768586863","https://openalex.org/W6768851824","https://openalex.org/W6769311223","https://openalex.org/W6769692749","https://openalex.org/W6771175319","https://openalex.org/W6771713106","https://openalex.org/W6773813506","https://openalex.org/W6775706467","https://openalex.org/W6778883912","https://openalex.org/W6780226713","https://openalex.org/W6784072866","https://openalex.org/W6784698631"],"related_works":["https://openalex.org/W2058965144","https://openalex.org/W2164382479","https://openalex.org/W2146343568","https://openalex.org/W98480971","https://openalex.org/W2150291671","https://openalex.org/W2013643406","https://openalex.org/W2027972911","https://openalex.org/W2157978810","https://openalex.org/W2597809628","https://openalex.org/W4321593827"],"abstract_inverted_index":{"Large-scale":[0],"pre-trained":[1,54,96],"language":[2,21,36,62,97],"models":[3,56,90,98],"(e.g.,":[4],"BERT)":[5],"have":[6],"attracted":[7],"great":[8],"attention":[9],"in":[10,110,133],"recent":[11],"years.":[12],"It":[13],"is":[14,146],"straightforward":[15],"to":[16,126,148,153,211],"fine-tune":[17],"them":[18,33],"on":[19,82,169],"natural":[20,35],"understanding":[22],"tasks":[23,38],"such":[24,39],"as":[25,40],"text":[26],"classification,":[27],"however,":[28],"effectively":[29],"and":[30,60,80,161],"efficiently":[31],"incorporating":[32],"into":[34,64,105],"generation":[37],"neural":[41,171],"machine":[42,172],"translation":[43,173],"remains":[44],"a":[45,65,111],"challenging":[46],"problem.":[47],"In":[48,141],"this":[49,120,142],"paper,":[50],"we":[51,122],"integrate":[52],"two":[53],"BERT":[55,78,89],"from":[57],"the":[58,86,144,158,166,188,197,214,223],"source":[59],"target":[61],"domains":[63],"sequence-to-sequence":[66],"model":[67,145,181,199,216],"by":[68,191],"introducing":[69],"light-weight":[70],"adapter":[71,195],"modules.":[72],"The":[73],"adapters":[74,104,130,152],"are":[75,91,99,137],"inserted":[76],"between":[77],"layers":[79,107],"tuned":[81],"downstream":[83],"tasks,":[84],"while":[85,186,204],"parameters":[87],"of":[88,114],"fixed":[92],"during":[93,139],"fine-tuning.":[94,140],"As":[95],"usually":[100],"very":[101],"deep,":[102],"inserting":[103],"all":[106],"will":[108],"result":[109],"considerable":[112],"scale":[113],"new":[115],"parameters.":[116],"To":[117],"deal":[118],"with":[119,176,222],"problem,":[121],"introduce":[123],"latent":[124],"variables":[125],"decide":[127],"whether":[128],"using":[129],"or":[131],"not":[132],"each":[134],"layer,":[135],"which":[136,151],"learned":[138],"way,":[143],"able":[147],"automatically":[149],"determine":[150],"use,":[154],"therefore":[155],"hugely":[156],"promoting":[157],"parameter":[159],"efficiency":[160],"decoding":[162],"speed.":[163],"We":[164],"evaluate":[165],"proposed":[167,198,215],"framework":[168],"various":[170],"tasks.":[174],"Equipped":[175],"parallel":[177],"sequence":[178],"decoding,":[179,213],"our":[180],"consistently":[182],"outperforms":[183],"autoregressive":[184,207,212],"baselines":[185],"reducing":[187],"inference":[189],"latency":[190],"half.":[192],"With":[193],"automatic":[194],"selection,":[196],"further":[200],"achieves":[201],"20%":[202],"speedup":[203],"still":[205],"outperforming":[206],"baselines.":[208],"When":[209],"applied":[210],"can":[217],"also":[218],"achieve":[219],"comparable":[220],"performance":[221],"state-of-the-art":[224],"baseline":[225],"models.":[226]},"counts_by_year":[{"year":2026,"cited_by_count":3},{"year":2025,"cited_by_count":8},{"year":2024,"cited_by_count":9},{"year":2023,"cited_by_count":3},{"year":2022,"cited_by_count":4},{"year":2021,"cited_by_count":3}],"updated_date":"2026-04-11T08:14:18.477133","created_date":"2025-10-10T00:00:00"}