{"id":"https://openalex.org/W3201624868","doi":"https://doi.org/10.48550/arxiv.2109.08040","title":"Locating Language-Specific Information in Contextualized Embeddings","display_name":"Locating Language-Specific Information in Contextualized Embeddings","publication_year":2021,"publication_date":"2021-09-16","ids":{"openalex":"https://openalex.org/W3201624868","doi":"https://doi.org/10.48550/arxiv.2109.08040","mag":"3201624868"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2109.08040","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2109.08040","pdf_url":"https://arxiv.org/pdf/2109.08040","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"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":"text"},"volume":null,"type":"preprint","indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2109.08040","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5110721336","display_name":"Sheng Liang","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I8204097","display_name":"Ludwig-Maximilians-Universit\u00e4t M\u00fcnchen","ror":"https://ror.org/05591te55","country_code":"DE","type":"education","lineage":["https://openalex.org/I8204097"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Liang, Sheng","raw_affiliation_strings":["Center for Information and Language Processing (CIS) LMU Munich , Germany"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Center for Information and Language Processing (CIS) LMU Munich , Germany","institution_ids":["https://openalex.org/I8204097"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5012225484","display_name":"Philipp Dufter","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I8204097","display_name":"Ludwig-Maximilians-Universit\u00e4t M\u00fcnchen","ror":"https://ror.org/05591te55","country_code":"DE","type":"education","lineage":["https://openalex.org/I8204097"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Dufter, Philipp","raw_affiliation_strings":["Center for Information and Language Processing (CIS) LMU Munich , Germany"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Center for Information and Language Processing (CIS) LMU Munich , Germany","institution_ids":["https://openalex.org/I8204097"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5071144367","display_name":"Hinrich Sch\u00fctze","orcid":"https://orcid.org/0000-0001-9514-7934","observed_orcids":["https://orcid.org/0000-0001-9514-7934"]},"institutions":[{"id":"https://openalex.org/I8204097","display_name":"Ludwig-Maximilians-Universit\u00e4t M\u00fcnchen","ror":"https://ror.org/05591te55","country_code":"DE","type":"education","lineage":["https://openalex.org/I8204097"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Sch\u00fctze, Hinrich","raw_affiliation_strings":["Center for Information and Language Processing (CIS) LMU Munich , Germany"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Center for Information and Language Processing (CIS) LMU Munich , Germany","institution_ids":["https://openalex.org/I8204097"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I8204097"],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"cited_by_count":4,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":null,"last_page":null},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.9692000150680542,"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":0.9692000150680542,"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":0.01510000042617321,"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/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.007000000216066837,"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/multilingual-pretrained-language-models","display_name":"multilingual pretrained language models","score":0.9959999918937683},{"id":"https://openalex.org/keywords/contextualized-embeddings","display_name":"contextual embeddings","score":0.9079999923706055},{"id":"https://openalex.org/keywords/multilingualism","display_name":"multilingualism","score":0.7860000133514404},{"id":"https://openalex.org/keywords/representation-analysis","display_name":"representation analysis","score":0.5479999780654907},{"id":"https://openalex.org/keywords/layer-by-layer-analysis","display_name":"layer-wise analysis","score":0.5289999842643738},{"id":"https://openalex.org/keywords/cross-lingual-transfer","display_name":"cross-lingual transfer","score":0.6589999794960022},{"id":"https://openalex.org/keywords/unsupervised-pretraining","display_name":"unsupervised pre-training","score":0.6190000176429749}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7452319860458374},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.6572768092155457},{"id":"https://openalex.org/C111030470","wikidata":"https://www.wikidata.org/wiki/Q1430460","display_name":"Curse of dimensionality","level":2,"score":0.6327452063560486},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.6163350343704224},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.603263258934021},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.5695056915283203},{"id":"https://openalex.org/C32834561","wikidata":"https://www.wikidata.org/wiki/Q660730","display_name":"Subspace topology","level":2,"score":0.5014863014221191},{"id":"https://openalex.org/C187736073","wikidata":"https://www.wikidata.org/wiki/Q2920921","display_name":"Management","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":2,"locations":[{"id":"pmh:oai:arXiv.org:2109.08040","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2109.08040","pdf_url":"https://arxiv.org/pdf/2109.08040","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"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":"text"},{"id":"doi:10.48550/arxiv.2109.08040","doi":"https://doi.org/10.48550/arxiv.2109.08040","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2109.08040","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2109.08040","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2109.08040","pdf_url":"https://arxiv.org/pdf/2109.08040","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"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":"text"},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":15,"referenced_works":["https://openalex.org/W2001619934","https://openalex.org/W2152939563","https://openalex.org/W2286496058","https://openalex.org/W2936032120","https://openalex.org/W2963341956","https://openalex.org/W2970854433","https://openalex.org/W2983040767","https://openalex.org/W3015299511","https://openalex.org/W3035547806","https://openalex.org/W3087873698","https://openalex.org/W3092858501","https://openalex.org/W3122044994","https://openalex.org/W3170488153","https://openalex.org/W3186903869","https://openalex.org/W6893675597"],"related_works":["https://openalex.org/W6990139777","https://openalex.org/W4385573436","https://openalex.org/W7169773329","https://openalex.org/W7079711872","https://openalex.org/W7168356155","https://openalex.org/W7166799070","https://openalex.org/W4385572966","https://openalex.org/W7223033859","https://openalex.org/W4412945246","https://openalex.org/W7126465701"],"abstract_inverted_index":{"Multilingual":[0],"pretrained":[1],"language":[2],"models":[3],"(MPLMs)":[4],"exhibit":[5],"multilinguality":[6,30],"and":[7,23,29,62,66],"are":[8,17,40],"well":[9,46],"suited":[10],"for":[11],"transfer":[12],"across":[13,80],"languages.":[14],"Most":[15],"MPLMs":[16,61],"trained":[18],"in":[19,60],"an":[20],"unsupervised":[21],"fashion":[22],"the":[24,35,67],"relationship":[25],"between":[26],"their":[27],"objective":[28],"is":[31,78],"unclear.":[32],"More":[33],"specifically,":[34],"question":[36],"whether":[37],"MPLM":[38,99],"representations":[39],"language-agnostic":[41],"or":[42],"they":[43],"simply":[44],"interleave":[45],"with":[47],"learned":[48],"task":[49],"prediction":[50],"heads":[51],"arises.":[52],"In":[53],"this":[54,70],"work,":[55],"we":[56],"locate":[57],"language-specific":[58,76],"information":[59,71,77],"identify":[63],"its":[64],"dimensionality":[65],"layers":[68],"where":[69],"occurs.":[72],"We":[73],"show":[74],"that":[75],"scattered":[79],"many":[81],"dimensions,":[82],"which":[83],"can":[84],"be":[85],"projected":[86],"into":[87],"a":[88,95],"linear":[89],"subspace.":[90],"Our":[91],"study":[92],"contributes":[93],"to":[94],"better":[96],"understanding":[97],"of":[98,108],"representations,":[100],"going":[101],"beyond":[102],"treating":[103],"them":[104],"as":[105],"unanalyzable":[106],"blobs":[107],"information.":[109]},"counts_by_year":[{"year":2023,"cited_by_count":2},{"year":2022,"cited_by_count":2}],"updated_date":"2026-10-16T09:59:22.235173","created_date":"2021-09-27T00:00:00"}