{"id":"https://openalex.org/W7166848261","doi":"https://doi.org/10.18653/v1/2026.acl-long.1554","title":"Multilingual Language Models Encode Script Over Linguistic Structure","display_name":"Multilingual Language Models Encode Script Over Linguistic Structure","publication_year":2026,"publication_date":"2026-01-01","ids":{"openalex":"https://openalex.org/W7166848261","doi":"https://doi.org/10.18653/v1/2026.acl-long.1554"},"language":null,"primary_location":{"id":"doi:10.18653/v1/2026.acl-long.1554","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.acl-long.1554","pdf_url":"https://aclanthology.org/2026.acl-long.1554.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 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://aclanthology.org/2026.acl-long.1554.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5133204861","display_name":"Aastha A K Verma","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Aastha A K Verma","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5104707682","display_name":"Anwoy Chatterjee","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Anwoy Chatterjee","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139731923","display_name":"Mehak Gupta","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Mehak Gupta","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5139822722","display_name":"Tanmoy Chakraborty","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Tanmoy Chakraborty","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":true,"cited_by_count":0,"citation_normalized_percentile":{"value":0.83806788,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"33685","last_page":"33719"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10181","display_name":"Natural Language Processing Techniques","score":0.45329999923706055,"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":0.45329999923706055,"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/T12090","display_name":"Language and cultural evolution","score":0.09839999675750732,"subfield":{"id":"https://openalex.org/subfields/3316","display_name":"Cultural Studies"},"field":{"id":"https://openalex.org/fields/33","display_name":"Social Sciences"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}},{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.09539999812841415,"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/encode","display_name":"ENCODE","score":0.49160000681877136},{"id":"https://openalex.org/keywords/information-structure","display_name":"Information structure","score":0.3517000079154968},{"id":"https://openalex.org/keywords/feature","display_name":"Feature (linguistics)","score":0.3009999990463257},{"id":"https://openalex.org/keywords/natural-language","display_name":"Natural language","score":0.29499998688697815},{"id":"https://openalex.org/keywords/semantics","display_name":"Semantics (computer science)","score":0.2816999852657318}],"concepts":[{"id":"https://openalex.org/C41895202","wikidata":"https://www.wikidata.org/wiki/Q8162","display_name":"Linguistics","level":1,"score":0.6802999973297119},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6394000053405762},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.5478000044822693},{"id":"https://openalex.org/C66746571","wikidata":"https://www.wikidata.org/wiki/Q1134833","display_name":"ENCODE","level":3,"score":0.49160000681877136},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4474000036716461},{"id":"https://openalex.org/C2779954242","wikidata":"https://www.wikidata.org/wiki/Q6031227","display_name":"Information structure","level":2,"score":0.3517000079154968},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.3009999990463257},{"id":"https://openalex.org/C195324797","wikidata":"https://www.wikidata.org/wiki/Q33742","display_name":"Natural language","level":2,"score":0.29499998688697815},{"id":"https://openalex.org/C184337299","wikidata":"https://www.wikidata.org/wiki/Q1437428","display_name":"Semantics (computer science)","level":2,"score":0.2816999852657318},{"id":"https://openalex.org/C2780035574","wikidata":"https://www.wikidata.org/wiki/Q30081","display_name":"Multilingualism","level":2,"score":0.26190000772476196},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.2596000134944916}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.18653/v1/2026.acl-long.1554","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.acl-long.1554","pdf_url":"https://aclanthology.org/2026.acl-long.1554.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 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.18653/v1/2026.acl-long.1554","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.acl-long.1554","pdf_url":"https://aclanthology.org/2026.acl-long.1554.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 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)","raw_type":"proceedings-article"},"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/4","score":0.651604413986206,"display_name":"Quality Education"}],"awards":[],"funders":[{"id":"https://openalex.org/F4320306076","display_name":"National Science Foundation","ror":"https://ror.org/021nxhr62"},{"id":"https://openalex.org/F4320320671","display_name":"National Research Foundation","ror":"https://ror.org/05s0g1g46"},{"id":"https://openalex.org/F4320320719","display_name":"Department of Science and Technology, Ministry of Science and Technology, India","ror":"https://ror.org/0101xrq71"}],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7166848261.pdf","grobid_xml":"https://content.openalex.org/works/W7166848261.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Multilingual":[0],"language":[1,34],"models":[2],"(LMs)":[3],"organize":[4,141],"representations":[5,81,142],"for":[6],"typologically":[7],"and":[8,52,62],"orthographically":[9],"diverse":[10],"languages":[11],"into":[12,153],"a":[13,154],"shared":[14],"parameter":[15],"space,":[16],"yet":[17],"the":[18,55],"nature":[19],"of":[20],"this":[21,26],"internal":[22],"organization":[23],"remains":[24],"elusive.In":[25],"work,":[27],"we":[28,44],"investigate":[29],"which":[30],"linguistic":[31,147],"properties":[32],"-abstract":[33],"identity":[35],"or":[36],"surface-form":[37,124],"cues":[38],"-shape":[39],"multilingual":[40,139],"representations.To":[41],"do":[42],"so,":[43],"analyze":[45],"language-associated":[46],"units":[47,72,119,129],"across":[48],"different":[49],"model":[50],"families":[51],"scales":[53],"using":[54],"Language":[56],"Activation":[57],"Probability":[58],"Entropy":[59],"(LAPE)":[60],"metric,":[61],"further":[63],"decompose":[64],"activations":[65],"with":[66,84,146],"Sparse":[67],"Autoencoders.We":[68],"find":[69],"that":[70,82,100,113,120,138],"these":[71],"are":[73,121],"strongly":[74],"conditioned":[75],"on":[76,96],"orthography:":[77],"romanization":[78],"induces":[79],"near-disjoint":[80],"align":[83],"neither":[85],"native-script":[86],"inputs":[87],"nor":[88],"English,":[89],"while":[90,109],"word-order":[91],"shuffling":[92],"has":[93],"limited":[94],"effect":[95],"unit":[97],"identity.Probing":[98],"shows":[99],"typological":[101,132],"structure":[102],"becomes":[103],"increasingly":[104],"accessible":[105],"in":[106],"deeper":[107],"layers,":[108],"causal":[110],"interventions":[111],"indicate":[112],"generation":[114],"is":[115],"most":[116],"sensitive":[117],"to":[118,123,128],"invariant":[122],"perturbations":[125],"rather":[126],"than":[127],"identified":[130],"by":[131],"alignment":[133],"alone.Overall,":[134],"our":[135],"results":[136],"suggest":[137],"LMs":[140],"around":[143],"surface":[144],"form,":[145],"abstraction":[148],"emerging":[149],"gradually":[150],"without":[151],"collapsing":[152],"unified":[155],"interlingua.":[156]},"counts_by_year":[],"updated_date":"2026-07-29T14:22:42.915294","created_date":"2026-07-02T00:00:00"}