{"id":"https://openalex.org/W4415327347","doi":"https://doi.org/10.48550/arxiv.2505.14990","title":"Language Specific Knowledge: Do Models Know Better in X than in English?","display_name":"Language Specific Knowledge: Do Models Know Better in X than in English?","publication_year":2025,"publication_date":"2025-05-21","ids":{"openalex":"https://openalex.org/W4415327347","doi":"https://doi.org/10.48550/arxiv.2505.14990"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2505.14990","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2505.14990","pdf_url":"https://arxiv.org/pdf/2505.14990","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":"cc-by-nc-sa","license_id":"https://openalex.org/licenses/cc-by-nc-sa","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/2505.14990","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5099230662","display_name":"Ishika Agarwal","orcid":"https://orcid.org/0009-0002-7249-3847","observed_orcids":["https://orcid.org/0009-0002-7249-3847"]},"institutions":[{"id":"https://openalex.org/I157725225","display_name":"University of Illinois Urbana-Champaign","ror":"https://ror.org/047426m28","country_code":"US","type":"education","lineage":["https://openalex.org/I157725225"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Agarwal, Ishika","raw_affiliation_strings":["Department of Computer Science University of Illinois , Urbana-Champaign"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Department of Computer Science University of Illinois , Urbana-Champaign","institution_ids":["https://openalex.org/I157725225"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5092596147","display_name":"Nimet Beyza Bozdag","orcid":"https://orcid.org/0009-0009-8078-3562","observed_orcids":["https://orcid.org/0009-0009-8078-3562"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Bozdag, Nimet Beyza","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5152927330","display_name":"Dilek Hakkani-T\u00fcr","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Hakkani-T\u00fcr, Dilek","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"cited_by_count":0,"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":0.9682999849319458,"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":0.9682999849319458,"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":0.021700000390410423,"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":0.004600000102072954,"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/language-choice","display_name":"language choice","score":0.7210000157356262},{"id":"https://openalex.org/keywords/multilingual-language-models","display_name":"multilingual language models","score":0.8209999799728394},{"id":"https://openalex.org/keywords/question-answering","display_name":"question answering","score":0.8920000195503235},{"id":"https://openalex.org/keywords/query-languages","display_name":"query languages","score":0.6359999775886536},{"id":"https://openalex.org/keywords/low-resource-languages","display_name":"low-resource languages","score":0.7820000052452087},{"id":"https://openalex.org/keywords/sociocultural-norms","display_name":"sociocultural norms","score":0.6489999890327454}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7875999808311462},{"id":"https://openalex.org/C165696696","wikidata":"https://www.wikidata.org/wiki/Q11287","display_name":"Exploit","level":2,"score":0.6212999820709229},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5993000268936157},{"id":"https://openalex.org/C81917197","wikidata":"https://www.wikidata.org/wiki/Q628760","display_name":"Selection (genetic algorithm)","level":2,"score":0.582099974155426},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.5814999938011169},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.5737000107765198},{"id":"https://openalex.org/C129792486","wikidata":"https://www.wikidata.org/wiki/Q1050419","display_name":"Language identification","level":3,"score":0.5351999998092651},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.5073999762535095},{"id":"https://openalex.org/C2777267654","wikidata":"https://www.wikidata.org/wiki/Q3519023","display_name":"Test (biology)","level":2,"score":0.4465000033378601},{"id":"https://openalex.org/C83479923","wikidata":"https://www.wikidata.org/wiki/Q2063748","display_name":"Universal Networking Language","level":4,"score":0.4440000057220459},{"id":"https://openalex.org/C44291984","wikidata":"https://www.wikidata.org/wiki/Q1074173","display_name":"Question answering","level":2,"score":0.4291999936103821},{"id":"https://openalex.org/C2780586882","wikidata":"https://www.wikidata.org/wiki/Q7520643","display_name":"Simple (philosophy)","level":2,"score":0.39719998836517334},{"id":"https://openalex.org/C94922259","wikidata":"https://www.wikidata.org/wiki/Q33215","display_name":"Constructed language","level":2,"score":0.3792000114917755},{"id":"https://openalex.org/C195324797","wikidata":"https://www.wikidata.org/wiki/Q33742","display_name":"Natural language","level":2,"score":0.3677000105381012},{"id":"https://openalex.org/C136197465","wikidata":"https://www.wikidata.org/wiki/Q1729295","display_name":"Variety (cybernetics)","level":2,"score":0.34209999442100525},{"id":"https://openalex.org/C9652623","wikidata":"https://www.wikidata.org/wiki/Q190109","display_name":"Field (mathematics)","level":2,"score":0.33799999952316284},{"id":"https://openalex.org/C184337299","wikidata":"https://www.wikidata.org/wiki/Q1437428","display_name":"Semantics (computer science)","level":2,"score":0.31220000982284546},{"id":"https://openalex.org/C185954173","wikidata":"https://www.wikidata.org/wiki/Q4347052","display_name":"Object language","level":3,"score":0.3109000027179718},{"id":"https://openalex.org/C61797465","wikidata":"https://www.wikidata.org/wiki/Q1188986","display_name":"Term (time)","level":2,"score":0.29739999771118164},{"id":"https://openalex.org/C39608478","wikidata":"https://www.wikidata.org/wiki/Q5015979","display_name":"Cache language model","level":5,"score":0.2759999930858612},{"id":"https://openalex.org/C2779313563","wikidata":"https://www.wikidata.org/wiki/Q17072565","display_name":"On Language","level":2,"score":0.25519999861717224},{"id":"https://openalex.org/C74672266","wikidata":"https://www.wikidata.org/wiki/Q815859","display_name":"Language acquisition","level":2,"score":0.2508000135421753},{"id":"https://openalex.org/C32977378","wikidata":"https://www.wikidata.org/wiki/Q604737","display_name":"Data control language","level":5,"score":0.25040000677108765}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2505.14990","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2505.14990","pdf_url":"https://arxiv.org/pdf/2505.14990","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":"cc-by-nc-sa","license_id":"https://openalex.org/licenses/cc-by-nc-sa","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2505.14990","doi":"https://doi.org/10.48550/arxiv.2505.14990","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2505.14990","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":"cc-by-nc-sa","license_id":"https://openalex.org/licenses/cc-by-nc-sa","version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2505.14990","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2505.14990","pdf_url":"https://arxiv.org/pdf/2505.14990","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":"cc-by-nc-sa","license_id":"https://openalex.org/licenses/cc-by-nc-sa","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[],"sustainable_development_goals_aurora":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4415327347.pdf","grobid_xml":"https://content.openalex.org/works/W4415327347.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W7163872429","https://openalex.org/W7166816047","https://openalex.org/W4411113080","https://openalex.org/W7215801627","https://openalex.org/W4391555728","https://openalex.org/W4416078712","https://openalex.org/W7125530491","https://openalex.org/W7113452018","https://openalex.org/W6964415693","https://openalex.org/W7166891439"],"abstract_inverted_index":{"Often,":[0],"multilingual":[1],"language":[2,39,52,93,99,126,144,178,186,230],"models":[3,100,199,210,231],"are":[4,72,233,247],"trained":[5],"with":[6,238],"the":[7,18,38,41,47,62,90,119,124,128,143,173,182,190,226,239],"objective":[8],"to":[9,68,122,137,140,225],"map":[10,193],"semantically":[11],"similar":[12],"content":[13],"(in":[14],"different":[15],"languages)":[16],"in":[17,29,75,106,114,207,217,244],"same":[19],"latent":[20],"space.":[21],"In":[22],"this":[23,30],"paper,":[24],"we":[25,44,60,131,158],"show":[26,175],"a":[27,80,133,185],"nuance":[28],"training":[31],"objective,":[32],"and":[33,118,168,188,204,215,219,235,241],"find":[34],"that":[35,71,162,176,189,232],"by":[36],"changing":[37],"of":[40,51,92,135,149,184,229],"input":[42],"query,":[43],"can":[45,101,180],"improve":[46,181],"question":[48],"answering":[49],"ability":[50],"models.":[53],"We":[54,88],"make":[55],"two":[56],"main":[57],"contributions.":[58],"First,":[59],"introduce":[61,89,132],"term":[63],"Language":[64],"Specific":[65],"Knowledge":[66],"(LSK)":[67],"denote":[69],"queries":[70],"best":[73],"answered":[74],"an":[76],"``expert":[77],"language''":[78],"for":[79,96,127],"given":[81],"LLM,":[82],"thereby":[83],"enhancing":[84],"its":[85],"question-answering":[86],"ability.":[87],"problem":[91,146],"selection":[94,145,179],"--":[95,117],"some":[97],"queries,":[98],"perform":[102],"better":[103,113],"when":[104],"queried":[105],"languages":[107,116],"other":[108],"than":[109],"English,":[110],"sometimes":[111],"even":[112],"low-resource":[115],"goal":[120],"is":[121,194],"select":[123],"optimal":[125],"query.":[129],"Second,":[130],"variety":[134],"simple":[136],"strong":[138],"baselines":[139],"empirically":[141],"motivate":[142],"(including":[147],"one":[148],"our":[150,156,222],"own":[151],"methods":[152],"called":[153],"LSKExtractor).":[154],"During":[155],"evaluation,":[157],"employ":[159],"three":[160],"datasets":[161],"contain":[163],"knowledge":[164],"about":[165,202,213],"both":[166],"cultural":[167,240],"social":[169],"behavioral":[170],"norms.":[171],"Overall,":[172],"results":[174],"principled":[177],"performance":[183],"model,":[187],"expected":[191],"question-to-language":[192],"not":[195],"always":[196],"intuitive:":[197],"Gemma":[198],"know":[200,211],"most":[201,212],"China":[203],"Middle":[205],"East":[206],"Spanish;":[208],"Qwen":[209],"authority":[214],"responsibility":[216],"Arabic":[218],"Chinese.":[220],"Broadly,":[221],"research":[223],"contributes":[224],"open-source":[227],"development":[228],"inclusive":[234],"more":[236],"aligned":[237],"linguistic":[242],"contexts":[243],"which":[245],"they":[246],"deployed.":[248]},"counts_by_year":[],"updated_date":"2026-09-30T09:27:27.532063","created_date":"2025-10-19T00:00:00"}