{"id":"https://openalex.org/W2912351236","doi":"https://doi.org/10.18653/v1/w18-5431","title":"An Analysis of Encoder Representations in Transformer-Based Machine Translation","display_name":"An Analysis of Encoder Representations in Transformer-Based Machine Translation","publication_year":2018,"publication_date":"2018-01-01","ids":{"openalex":"https://openalex.org/W2912351236","doi":"https://doi.org/10.18653/v1/w18-5431","mag":"2912351236"},"language":"en","primary_location":{"id":"doi:10.18653/v1/w18-5431","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/w18-5431","pdf_url":"https://www.aclweb.org/anthology/W18-5431.pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 2018 EMNLP Workshop BlackboxNLP: Analyzing and Interpreting Neural Networks for NLP","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://www.aclweb.org/anthology/W18-5431.pdf","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5053806445","display_name":"Alessandro Raganato","orcid":"https://orcid.org/0000-0002-7018-7515"},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Alessandro Raganato","raw_affiliation_strings":["Department of Digital Humanities University of Helsinki"],"affiliations":[{"raw_affiliation_string":"Department of Digital Humanities University of Helsinki","institution_ids":[]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5082417280","display_name":"J\u00f6rg Tiedemann","orcid":"https://orcid.org/0000-0003-3065-7989"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"J\u00f6rg Tiedemann","raw_affiliation_strings":["Department of Digital Humanities University of Helsinki"],"affiliations":[{"raw_affiliation_string":"Department of Digital Humanities University of Helsinki","institution_ids":[]}]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":2,"corresponding_author_ids":["https://openalex.org/A5053806445"],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":26.7065,"has_fulltext":true,"cited_by_count":269,"citation_normalized_percentile":{"value":0.99588957,"is_in_top_1_percent":true,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":90,"max":100},"biblio":{"volume":null,"issue":null,"first_page":"287","last_page":"297"},"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":0.9998999834060669,"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":0.9865000247955322,"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":0.842962384223938},{"id":"https://openalex.org/keywords/machine-translation","display_name":"Machine translation","score":0.8054342269897461},{"id":"https://openalex.org/keywords/transformer","display_name":"Transformer","score":0.7853385210037231},{"id":"https://openalex.org/keywords/encoder","display_name":"Encoder","score":0.6473790407180786},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.635279655456543},{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.6144518256187439},{"id":"https://openalex.org/keywords/syntax","display_name":"Syntax","score":0.5069491863250732},{"id":"https://openalex.org/keywords/encode","display_name":"ENCODE","score":0.5048361420631409},{"id":"https://openalex.org/keywords/dependency","display_name":"Dependency (UML)","score":0.47245004773139954}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.842962384223938},{"id":"https://openalex.org/C203005215","wikidata":"https://www.wikidata.org/wiki/Q79798","display_name":"Machine translation","level":2,"score":0.8054342269897461},{"id":"https://openalex.org/C66322947","wikidata":"https://www.wikidata.org/wiki/Q11658","display_name":"Transformer","level":3,"score":0.7853385210037231},{"id":"https://openalex.org/C118505674","wikidata":"https://www.wikidata.org/wiki/Q42586063","display_name":"Encoder","level":2,"score":0.6473790407180786},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.635279655456543},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.6144518256187439},{"id":"https://openalex.org/C60048249","wikidata":"https://www.wikidata.org/wiki/Q37437","display_name":"Syntax","level":2,"score":0.5069491863250732},{"id":"https://openalex.org/C66746571","wikidata":"https://www.wikidata.org/wiki/Q1134833","display_name":"ENCODE","level":3,"score":0.5048361420631409},{"id":"https://openalex.org/C19768560","wikidata":"https://www.wikidata.org/wiki/Q320727","display_name":"Dependency (UML)","level":2,"score":0.47245004773139954},{"id":"https://openalex.org/C55493867","wikidata":"https://www.wikidata.org/wiki/Q7094","display_name":"Biochemistry","level":1,"score":0.0},{"id":"https://openalex.org/C62520636","wikidata":"https://www.wikidata.org/wiki/Q944","display_name":"Quantum mechanics","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/C165801399","wikidata":"https://www.wikidata.org/wiki/Q25428","display_name":"Voltage","level":2,"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/C104317684","wikidata":"https://www.wikidata.org/wiki/Q7187","display_name":"Gene","level":2,"score":0.0},{"id":"https://openalex.org/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","level":0,"score":0.0}],"mesh":[],"locations_count":4,"locations":[{"id":"doi:10.18653/v1/w18-5431","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/w18-5431","pdf_url":"https://www.aclweb.org/anthology/W18-5431.pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 2018 EMNLP Workshop BlackboxNLP: Analyzing and Interpreting Neural Networks for NLP","raw_type":"proceedings-article"},{"id":"pmh:oai:boa.unimib.it:10281/361565","is_oa":false,"landing_page_url":"http://hdl.handle.net/10281/361565","pdf_url":null,"source":{"id":"https://openalex.org/S4306401259","display_name":"BOA (University of Milano-Bicocca)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I66752286","host_organization_name":"University of Milano-Bicocca","host_organization_lineage":["https://openalex.org/I66752286"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"info:eu-repo/semantics/conferenceObject"},{"id":"pmh:oai:helda.helsinki.fi:10138/263704","is_oa":true,"landing_page_url":"http://hdl.handle.net/10138/263704","pdf_url":null,"source":{"id":"https://openalex.org/S4210213322","display_name":"Ty\u00f6v\u00e4entutkimus Vuosikirja","issn_l":"0784-1272","issn":["0784-1272","1459-7780"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"journal"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"","raw_type":"Conference contribution"},{"id":"pmh:oai:iris.uniroma1.it:11573/1553731","is_oa":false,"landing_page_url":"https://hdl.handle.net/11573/1553731","pdf_url":null,"source":{"id":"https://openalex.org/S4377196107","display_name":"IRIS Research product catalog (Sapienza University of Rome)","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":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"info:eu-repo/semantics/conferenceObject"}],"best_oa_location":{"id":"doi:10.18653/v1/w18-5431","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/w18-5431","pdf_url":"https://www.aclweb.org/anthology/W18-5431.pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 2018 EMNLP Workshop BlackboxNLP: Analyzing and Interpreting Neural Networks for NLP","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[],"funders":[{"id":"https://openalex.org/F4320309480","display_name":"Nvidia","ror":"https://ror.org/03jdj4y14"},{"id":"https://openalex.org/F4320321108","display_name":"Academy of Finland","ror":"https://ror.org/05k73zm37"}],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W2912351236.pdf","grobid_xml":"https://content.openalex.org/works/W2912351236.grobid-xml"},"referenced_works_count":65,"referenced_works":["https://openalex.org/W22168010","https://openalex.org/W630532510","https://openalex.org/W1689711448","https://openalex.org/W1902237438","https://openalex.org/W1924770834","https://openalex.org/W1951216520","https://openalex.org/W2087946919","https://openalex.org/W2098921539","https://openalex.org/W2118434577","https://openalex.org/W2130942839","https://openalex.org/W2133564696","https://openalex.org/W2140030083","https://openalex.org/W2157331557","https://openalex.org/W2167615167","https://openalex.org/W2250606284","https://openalex.org/W2292919134","https://openalex.org/W2512924740","https://openalex.org/W2515741950","https://openalex.org/W2525778437","https://openalex.org/W2549835527","https://openalex.org/W2552839021","https://openalex.org/W2554915555","https://openalex.org/W2560864221","https://openalex.org/W2563351168","https://openalex.org/W2563574619","https://openalex.org/W2594470997","https://openalex.org/W2605717780","https://openalex.org/W2613904329","https://openalex.org/W2625014264","https://openalex.org/W2739827909","https://openalex.org/W2740840489","https://openalex.org/W2741040846","https://openalex.org/W2752194699","https://openalex.org/W2757154661","https://openalex.org/W2760656271","https://openalex.org/W2773956126","https://openalex.org/W2791751435","https://openalex.org/W2799124508","https://openalex.org/W2885588803","https://openalex.org/W2951299559","https://openalex.org/W2952087486","https://openalex.org/W2952800841","https://openalex.org/W2962708992","https://openalex.org/W2962717763","https://openalex.org/W2962776659","https://openalex.org/W2962777840","https://openalex.org/W2962784628","https://openalex.org/W2962911926","https://openalex.org/W2962982474","https://openalex.org/W2963088995","https://openalex.org/W2963123635","https://openalex.org/W2963212250","https://openalex.org/W2963403868","https://openalex.org/W2963506925","https://openalex.org/W2963532001","https://openalex.org/W2963626623","https://openalex.org/W2963641561","https://openalex.org/W2963751529","https://openalex.org/W2963899396","https://openalex.org/W2964204621","https://openalex.org/W2964265128","https://openalex.org/W2964308564","https://openalex.org/W4248358431","https://openalex.org/W4299668319","https://openalex.org/W4385245566"],"related_works":["https://openalex.org/W2468279273","https://openalex.org/W2354198838","https://openalex.org/W1989130879","https://openalex.org/W2103419012","https://openalex.org/W2080244595","https://openalex.org/W4287125478","https://openalex.org/W3170826848","https://openalex.org/W3171594500","https://openalex.org/W2036499707","https://openalex.org/W2275988210"],"abstract_inverted_index":{"The":[0,16],"attention":[1,28,82,125,150],"mechanism":[2,83],"is":[3,24,51,78],"a":[4,127],"successful":[5],"technique":[6],"in":[7,11,38,84,126],"modern":[8],"NLP,":[9],"especially":[10],"tasks":[12,110],"like":[13],"machine":[14,40],"translation.":[15],"recently":[17],"proposed":[18],"network":[19],"architecture":[20],"of":[21,34,57,95,115,141],"the":[22,35,54,58,61,75,81,93,96,113,136,142],"Transformer":[23,85],"based":[25,102],"entirely":[26],"on":[27,103,135],"mechanisms":[29],"and":[30,60,117,121,139,156],"achieves":[31],"new":[32],"state":[33],"art":[36],"results":[37],"neural":[39],"translation,":[41],"outperforming":[42],"other":[43],"sequence-to-sequence":[44],"models.":[45],"However,":[46],"so":[47],"far":[48],"not":[49],"much":[50],"known":[52],"about":[53,168],"internal":[55],"properties":[56],"model":[59],"representations":[62,94],"it":[63],"learns":[64],"to":[65,111,165,174],"achieve":[66],"that":[67,77,148,161],"performance.":[68],"To":[69],"study":[70,112],"this":[71],"question,":[72],"we":[73,106,122,157],"investigate":[74],"information":[76,120],"learned":[79],"by":[80,98],"models":[86],"with":[87],"different":[88],"translation":[89],"quality.":[90],"We":[91,146],"assess":[92],"encoder":[97,144],"extracting":[99],"dependency":[100,154],"relations":[101,155],"self-attention":[104],"weights,":[105],"perform":[107],"four":[108],"probing":[109],"amount":[114],"syntactic":[116,153],"semantic":[118],"captured":[119],"also":[123,159],"test":[124],"transfer":[128],"learning":[129],"scenario.":[130],"Our":[131],"analysis":[132],"sheds":[133],"light":[134],"relative":[137],"strengths":[138],"weaknesses":[140],"various":[143],"representations.":[145],"observe":[147],"specific":[149],"heads":[151],"mark":[152],"can":[158],"confirm":[160],"lower":[162],"layers":[163,172],"tend":[164,173],"learn":[166],"more":[167,176],"syntax":[169],"while":[170],"higher":[171],"encode":[175],"semantics.":[177]},"counts_by_year":[{"year":2026,"cited_by_count":1},{"year":2025,"cited_by_count":17},{"year":2024,"cited_by_count":19},{"year":2023,"cited_by_count":36},{"year":2022,"cited_by_count":38},{"year":2021,"cited_by_count":51},{"year":2020,"cited_by_count":70},{"year":2019,"cited_by_count":36},{"year":2018,"cited_by_count":1}],"updated_date":"2026-04-04T16:13:02.066488","created_date":"2025-10-10T00:00:00"}