{"id":"https://openalex.org/W4394871646","doi":"https://doi.org/10.1162/tacl_a_00651","title":"Are Character-level Translations Worth the Wait? Comparing ByT5 and mT5 for Machine Translation","display_name":"Are Character-level Translations Worth the Wait? Comparing ByT5 and mT5 for Machine Translation","publication_year":2024,"publication_date":"2024-01-01","ids":{"openalex":"https://openalex.org/W4394871646","doi":"https://doi.org/10.1162/tacl_a_00651"},"language":"en","primary_location":{"id":"doi:10.1162/tacl_a_00651","is_oa":true,"landing_page_url":"https://doi.org/10.1162/tacl_a_00651","pdf_url":"https://direct.mit.edu/tacl/article-pdf/doi/10.1162/tacl_a_00651/2364129/tacl_a_00651.pdf","source":{"id":"https://openalex.org/S2729999759","display_name":"Transactions of the Association for Computational Linguistics","issn_l":"2307-387X","issn":["2307-387X"],"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","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Transactions of the Association for 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/tacl/article-pdf/doi/10.1162/tacl_a_00651/2364129/tacl_a_00651.pdf","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5055868817","display_name":"Lukas Edman","orcid":"https://orcid.org/0000-0001-8215-3614"},"institutions":[{"id":"https://openalex.org/I169381384","display_name":"University of Groningen","ror":"https://ror.org/012p63287","country_code":"NL","type":"education","lineage":["https://openalex.org/I169381384"]}],"countries":["NL"],"is_corresponding":true,"raw_author_name":"Lukas Edman","raw_affiliation_strings":["Center for Language and Cognition, University of Groningen, the Netherlands. j.l.edman@rug.nl"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Center for Language and Cognition, University of Groningen, the Netherlands. j.l.edman@rug.nl","institution_ids":["https://openalex.org/I169381384"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5042129410","display_name":"Gabriele Sarti","orcid":"https://orcid.org/0000-0001-8715-2987"},"institutions":[{"id":"https://openalex.org/I169381384","display_name":"University of Groningen","ror":"https://ror.org/012p63287","country_code":"NL","type":"education","lineage":["https://openalex.org/I169381384"]}],"countries":["NL"],"is_corresponding":true,"raw_author_name":"Gabriele Sarti","raw_affiliation_strings":["Center for Language and Cognition, University of Groningen, the Netherlands. g.sarti@rug.nl"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Center for Language and Cognition, University of Groningen, the Netherlands. g.sarti@rug.nl","institution_ids":["https://openalex.org/I169381384"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5062029765","display_name":"Antonio Toral","orcid":"https://orcid.org/0000-0003-2357-2960"},"institutions":[{"id":"https://openalex.org/I169381384","display_name":"University of Groningen","ror":"https://ror.org/012p63287","country_code":"NL","type":"education","lineage":["https://openalex.org/I169381384"]}],"countries":["NL"],"is_corresponding":true,"raw_author_name":"Antonio Toral","raw_affiliation_strings":["Center for Language and Cognition, University of Groningen, the Netherlands. a.toral.ruiz@rug.nl"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Center for Language and Cognition, University of Groningen, the Netherlands. a.toral.ruiz@rug.nl","institution_ids":["https://openalex.org/I169381384"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5047337772","display_name":"Gertjan van Noord","orcid":"https://orcid.org/0000-0001-5564-6341"},"institutions":[{"id":"https://openalex.org/I169381384","display_name":"University of Groningen","ror":"https://ror.org/012p63287","country_code":"NL","type":"education","lineage":["https://openalex.org/I169381384"]}],"countries":["NL"],"is_corresponding":true,"raw_author_name":"Gertjan van Noord","raw_affiliation_strings":["Center for Language and Cognition, University of Groningen, the Netherlands. g.j.m.van.noord@rug.nl"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Center for Language and Cognition, University of Groningen, the Netherlands. g.j.m.van.noord@rug.nl","institution_ids":["https://openalex.org/I169381384"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5019968969","display_name":"Arianna Bisazza","orcid":"https://orcid.org/0000-0003-1270-3048"},"institutions":[{"id":"https://openalex.org/I169381384","display_name":"University of Groningen","ror":"https://ror.org/012p63287","country_code":"NL","type":"education","lineage":["https://openalex.org/I169381384"]}],"countries":["NL"],"is_corresponding":true,"raw_author_name":"Arianna Bisazza","raw_affiliation_strings":["Center for Language and Cognition, University of Groningen, the Netherlands. a.bisazza@rug.nl"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Center for Language and Cognition, University of Groningen, the Netherlands. a.bisazza@rug.nl","institution_ids":["https://openalex.org/I169381384"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":["https://openalex.org/A5019968969","https://openalex.org/A5042129410","https://openalex.org/A5047337772","https://openalex.org/A5055868817","https://openalex.org/A5062029765"],"corresponding_institution_ids":["https://openalex.org/I169381384"],"apc_list":null,"apc_paid":null,"fwci":2.2228,"has_fulltext":true,"cited_by_count":8,"citation_normalized_percentile":{"value":0.88967422,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":90,"max":98},"biblio":{"volume":"12","issue":null,"first_page":"392","last_page":"410"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"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"}},"topics":[{"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/T10181","display_name":"Natural Language Processing Techniques","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":0.9839000105857849,"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":0.8577079772949219},{"id":"https://openalex.org/keywords/machine-translation","display_name":"Machine translation","score":0.8335774540901184},{"id":"https://openalex.org/keywords/character","display_name":"Character (mathematics)","score":0.7886418104171753},{"id":"https://openalex.org/keywords/byte","display_name":"Byte","score":0.639610767364502},{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.5994757413864136},{"id":"https://openalex.org/keywords/word","display_name":"Word (group theory)","score":0.572146475315094},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.5530050992965698},{"id":"https://openalex.org/keywords/translation","display_name":"Translation (biology)","score":0.5452668070793152},{"id":"https://openalex.org/keywords/language-model","display_name":"Language model","score":0.5031740069389343},{"id":"https://openalex.org/keywords/quality","display_name":"Quality (philosophy)","score":0.4668000340461731},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.385648250579834},{"id":"https://openalex.org/keywords/linguistics","display_name":"Linguistics","score":0.19635745882987976}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8577079772949219},{"id":"https://openalex.org/C203005215","wikidata":"https://www.wikidata.org/wiki/Q79798","display_name":"Machine translation","level":2,"score":0.8335774540901184},{"id":"https://openalex.org/C2780861071","wikidata":"https://www.wikidata.org/wiki/Q1062934","display_name":"Character (mathematics)","level":2,"score":0.7886418104171753},{"id":"https://openalex.org/C43364308","wikidata":"https://www.wikidata.org/wiki/Q8799","display_name":"Byte","level":2,"score":0.639610767364502},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.5994757413864136},{"id":"https://openalex.org/C90805587","wikidata":"https://www.wikidata.org/wiki/Q10944557","display_name":"Word (group theory)","level":2,"score":0.572146475315094},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5530050992965698},{"id":"https://openalex.org/C149364088","wikidata":"https://www.wikidata.org/wiki/Q185917","display_name":"Translation (biology)","level":4,"score":0.5452668070793152},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.5031740069389343},{"id":"https://openalex.org/C2779530757","wikidata":"https://www.wikidata.org/wiki/Q1207505","display_name":"Quality (philosophy)","level":2,"score":0.4668000340461731},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.385648250579834},{"id":"https://openalex.org/C41895202","wikidata":"https://www.wikidata.org/wiki/Q8162","display_name":"Linguistics","level":1,"score":0.19635745882987976},{"id":"https://openalex.org/C55493867","wikidata":"https://www.wikidata.org/wiki/Q7094","display_name":"Biochemistry","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/C104317684","wikidata":"https://www.wikidata.org/wiki/Q7187","display_name":"Gene","level":2,"score":0.0},{"id":"https://openalex.org/C111472728","wikidata":"https://www.wikidata.org/wiki/Q9471","display_name":"Epistemology","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/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.0},{"id":"https://openalex.org/C138885662","wikidata":"https://www.wikidata.org/wiki/Q5891","display_name":"Philosophy","level":0,"score":0.0},{"id":"https://openalex.org/C2524010","wikidata":"https://www.wikidata.org/wiki/Q8087","display_name":"Geometry","level":1,"score":0.0},{"id":"https://openalex.org/C105580179","wikidata":"https://www.wikidata.org/wiki/Q188928","display_name":"Messenger RNA","level":3,"score":0.0}],"mesh":[],"locations_count":4,"locations":[{"id":"doi:10.1162/tacl_a_00651","is_oa":true,"landing_page_url":"https://doi.org/10.1162/tacl_a_00651","pdf_url":"https://direct.mit.edu/tacl/article-pdf/doi/10.1162/tacl_a_00651/2364129/tacl_a_00651.pdf","source":{"id":"https://openalex.org/S2729999759","display_name":"Transactions of the Association for Computational Linguistics","issn_l":"2307-387X","issn":["2307-387X"],"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","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Transactions of the Association for Computational Linguistics","raw_type":"journal-article"},{"id":"pmh:oai:pure.rug.nl:openaire/05662e7a-ef21-4f72-a103-7d1a9da63807","is_oa":true,"landing_page_url":"https://research.rug.nl/en/publications/05662e7a-ef21-4f72-a103-7d1a9da63807","pdf_url":null,"source":{"id":"https://openalex.org/S4306400420","display_name":"University of Groningen research database (University of Groningen / Centre for Information Technology)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I169381384","host_organization_name":"University of Groningen","host_organization_lineage":["https://openalex.org/I169381384"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Edman, L, Sarti, G, Toral, A, van Noord, G & Bisazza, A 2024, 'Are Character-level Translations Worth the Wait? Comparing ByT5 and mT5 for Machine Translation', Transactions of the Association for Computational Linguistics, vol. 12, pp. 392-410. https://doi.org/10.1162/tacl_a_00651","raw_type":"info:eu-repo/semantics/article"},{"id":"pmh:oai:doaj.org/article:6a83d12d8d8a48e48bc6d9df2f23df8a","is_oa":true,"landing_page_url":"https://doaj.org/article/6a83d12d8d8a48e48bc6d9df2f23df8a","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":"Transactions of the Association for Computational Linguistics, Vol 12 (2024)","raw_type":"article"},{"id":"pmh:oai:pure.rug.nl:publications/05662e7a-ef21-4f72-a103-7d1a9da63807","is_oa":true,"landing_page_url":"https://hdl.handle.net/11370/05662e7a-ef21-4f72-a103-7d1a9da63807","pdf_url":null,"source":{"id":"https://openalex.org/S4306400420","display_name":"University of Groningen research database (University of Groningen / Centre for Information Technology)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I169381384","host_organization_name":"University of Groningen","host_organization_lineage":["https://openalex.org/I169381384"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"Edman, L, Sarti, G, Toral, A, van Noord, G & Bisazza, A 2024, 'Are Character-level Translations Worth the Wait? Comparing ByT5 and mT5 for Machine Translation', Transactions of the Association for Computational Linguistics, vol. 12, pp. 392-410. https://doi.org/10.1162/tacl_a_00651","raw_type":"info:eu-repo/semantics/article"}],"best_oa_location":{"id":"doi:10.1162/tacl_a_00651","is_oa":true,"landing_page_url":"https://doi.org/10.1162/tacl_a_00651","pdf_url":"https://direct.mit.edu/tacl/article-pdf/doi/10.1162/tacl_a_00651/2364129/tacl_a_00651.pdf","source":{"id":"https://openalex.org/S2729999759","display_name":"Transactions of the Association for Computational Linguistics","issn_l":"2307-387X","issn":["2307-387X"],"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","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Transactions of the Association for Computational Linguistics","raw_type":"journal-article"},"sustainable_development_goals":[{"display_name":"Quality Education","id":"https://metadata.un.org/sdg/4","score":0.7699999809265137}],"awards":[{"id":"https://openalex.org/G1650570625","display_name":"InDeep: Interpreting Deep Learning Models for Text and Sound","funder_award_id":"NWA.1292.19.399","funder_id":"https://openalex.org/F4320321800","funder_display_name":"Nederlandse Organisatie voor Wetenschappelijk Onderzoek"}],"funders":[{"id":"https://openalex.org/F4320320933","display_name":"Rijksuniversiteit Groningen","ror":"https://ror.org/012p63287"},{"id":"https://openalex.org/F4320321800","display_name":"Nederlandse Organisatie voor Wetenschappelijk Onderzoek","ror":"https://ror.org/04jsz6e67"}],"has_content":{"pdf":true,"grobid_xml":false},"content_urls":{"pdf":"https://content.openalex.org/works/W4394871646.pdf"},"referenced_works_count":49,"referenced_works":["https://openalex.org/W1981276685","https://openalex.org/W2259472270","https://openalex.org/W2531207078","https://openalex.org/W2733628661","https://openalex.org/W2757141711","https://openalex.org/W2758950307","https://openalex.org/W2766940977","https://openalex.org/W2767899794","https://openalex.org/W2885588803","https://openalex.org/W2913668833","https://openalex.org/W2946817437","https://openalex.org/W2951559648","https://openalex.org/W2962732637","https://openalex.org/W2962784628","https://openalex.org/W2962851944","https://openalex.org/W2963251942","https://openalex.org/W2963887123","https://openalex.org/W2971154170","https://openalex.org/W2971296520","https://openalex.org/W3001434439","https://openalex.org/W3105214104","https://openalex.org/W3115462295","https://openalex.org/W3135427360","https://openalex.org/W3155609600","https://openalex.org/W3164045210","https://openalex.org/W3169483174","https://openalex.org/W3173506780","https://openalex.org/W3173571438","https://openalex.org/W3174418826","https://openalex.org/W3187467055","https://openalex.org/W3197138502","https://openalex.org/W3199258042","https://openalex.org/W3207091856","https://openalex.org/W3213645763","https://openalex.org/W4232915137","https://openalex.org/W4285077564","https://openalex.org/W4288089799","https://openalex.org/W4376632433","https://openalex.org/W4385570354","https://openalex.org/W4385571586","https://openalex.org/W4385573199","https://openalex.org/W4385574037","https://openalex.org/W6612870711","https://openalex.org/W6680986496","https://openalex.org/W6692563993","https://openalex.org/W6746141323","https://openalex.org/W6751104502","https://openalex.org/W6754244489","https://openalex.org/W6769627184"],"related_works":["https://openalex.org/W121858127","https://openalex.org/W2775143306","https://openalex.org/W4289329995","https://openalex.org/W2373574911","https://openalex.org/W2046727832","https://openalex.org/W4247895841","https://openalex.org/W2757141711","https://openalex.org/W2883671469","https://openalex.org/W2728761353","https://openalex.org/W2944691285"],"abstract_inverted_index":{"Abstract":[0],"Pretrained":[1],"character-level":[2,74,134],"and":[3,54,59,64,107,133],"byte-level":[4],"language":[5],"models":[6,16,62],"have":[7],"been":[8,28],"shown":[9],"to":[10,130,154],"be":[11],"competitive":[12],"with":[13],"popular":[14,42],"subword":[15],"across":[17,51],"a":[18],"range":[19],"of":[20,57,73,103,114,145],"Natural":[21],"Language":[22],"Processing":[23],"tasks.":[24],"However,":[25],"there":[26],"has":[27],"little":[29],"research":[30],"on":[31,67],"their":[32,149],"effectiveness":[33,72],"for":[34],"neural":[35],"machine":[36],"translation":[37,96,156],"(NMT),":[38],"particularly":[39,78],"within":[40,125],"the":[41,71,112,142],"pretrain-then-finetune":[43],"paradigm.":[44],"This":[45],"work":[46],"performs":[47],"an":[48,128],"extensive":[49],"comparison":[50],"multiple":[52],"languages":[53],"experimental":[55],"conditions":[56],"character-":[58],"subword-level":[60],"pretrained":[61],"(ByT5":[63],"mT5,":[65],"respectively)":[66],"NMT.":[68],"We":[69,138],"show":[70,90],"modeling":[75],"in":[76,79,95,100,117,151],"translation,":[77],"cases":[80],"where":[81],"fine-tuning":[82],"data":[83],"is":[84],"limited.":[85],"In":[86],"our":[87],"analysis,":[88],"we":[89,121],"how":[91],"character":[92],"models\u2019":[93],"gains":[94],"quality":[97],"are":[98],"reflected":[99],"better":[101],"translations":[102],"orthographically":[104],"similar":[105],"words":[106],"rare":[108],"words.":[109],"While":[110],"evaluating":[111],"importance":[113],"source":[115],"texts":[116],"driving":[118],"model":[119],"predictions,":[120],"highlight":[122],"word-level":[123,132],"patterns":[124],"ByT5,":[126],"suggesting":[127,148],"ability":[129],"modulate":[131],"information":[135],"during":[136],"generation.":[137],"conclude":[139],"by":[140],"assessing":[141],"efficiency":[143],"tradeoff":[144],"byte":[146],"models,":[147],"usage":[150],"non-time-critical":[152],"scenarios":[153],"boost":[155],"quality.":[157]},"counts_by_year":[{"year":2026,"cited_by_count":2},{"year":2025,"cited_by_count":5},{"year":2024,"cited_by_count":1}],"updated_date":"2026-05-21T06:26:12.895304","created_date":"2025-10-10T00:00:00"}