{"id":"https://openalex.org/W7140073660","doi":"https://doi.org/10.18653/v1/2026.loreslm-1.40","title":"Pretraining and Benchmarking Modern Encoders for Latvian","display_name":"Pretraining and Benchmarking Modern Encoders for Latvian","publication_year":2026,"publication_date":"2026-01-01","ids":{"openalex":"https://openalex.org/W7140073660","doi":"https://doi.org/10.18653/v1/2026.loreslm-1.40"},"language":"en","primary_location":{"id":"doi:10.18653/v1/2026.loreslm-1.40","doi":"https://doi.org/10.18653/v1/2026.loreslm-1.40","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.loreslm-1.40","pdf_url":"https://aclanthology.org/2026.loreslm-1.40.pdf","source":{"id":"https://openalex.org/S7407107604","display_name":"Workshop on Language Models for Low-Resource Languages (LoResLM)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"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":"conference"},"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 Second Workshop on Language Models for Low-Resource Languages (LoResLM 2026)","raw_type":"proceedings-article"},"volume":null,"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"hybrid","oa_url":"https://aclanthology.org/2026.loreslm-1.40.pdf","any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5045959308","display_name":"Art\u016brs Znoti\u0146\u0161","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I4405255181","display_name":"UL Institute of Mathematics and Computer Science","ror":"https://ror.org/02q2vrq41","country_code":"LV","type":"facility","lineage":["https://openalex.org/I4405255181","https://openalex.org/I91123046"]},{"id":"https://openalex.org/I91123046","display_name":"University of Latvia","ror":"https://ror.org/05g3mes96","country_code":"LV","type":"education","lineage":["https://openalex.org/I91123046"]}],"countries":["LV"],"is_corresponding":true,"raw_author_name":"Arturs Znotins","raw_affiliation_strings":["Institute of Mathematics and Computer Science , University of Latvia 29","Raina bulv. , Riga , LV-1459 , Latvia"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Institute of Mathematics and Computer Science , University of Latvia 29","institution_ids":["https://openalex.org/I4405255181","https://openalex.org/I91123046"]},{"raw_affiliation_string":"Raina bulv. , Riga , LV-1459 , Latvia","institution_ids":[]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":["https://openalex.org/A5045959308"],"corresponding_institution_ids":["https://openalex.org/I4405255181","https://openalex.org/I91123046"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":true,"cited_by_count":0,"citation_normalized_percentile":{"value":0.13940809,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"461","last_page":"470"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.9286999702453613,"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":0.9286999702453613,"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":0.06729999929666519,"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":0.00039999998989515007,"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/latvian-language","display_name":"Latvian language","score":0.9710000157356262},{"id":"https://openalex.org/keywords/encoder-only-transformer","display_name":"encoder-only transformer","score":0.9229999780654907},{"id":"https://openalex.org/keywords/low-resource-languages","display_name":"low-resource languages","score":0.7379999756813049},{"id":"https://openalex.org/keywords/multilingual-pretraining","display_name":"multilingual pre-training","score":0.7009999752044678},{"id":"https://openalex.org/keywords/modernbert","display_name":"ModernBERT","score":0.9459999799728394},{"id":"https://openalex.org/keywords/roberta","display_name":"RoBERTa","score":0.9599999785423279}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6022999882698059},{"id":"https://openalex.org/C86251818","wikidata":"https://www.wikidata.org/wiki/Q816754","display_name":"Benchmarking","level":2,"score":0.5586000084877014},{"id":"https://openalex.org/C2780375031","wikidata":"https://www.wikidata.org/wiki/Q9078","display_name":"Latvian","level":2,"score":0.5019000172615051},{"id":"https://openalex.org/C118505674","wikidata":"https://www.wikidata.org/wiki/Q42586063","display_name":"Encoder","level":2,"score":0.45159998536109924},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.39809998869895935},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.2759000062942505},{"id":"https://openalex.org/C126042441","wikidata":"https://www.wikidata.org/wiki/Q1324888","display_name":"Frame (networking)","level":2,"score":0.27570000290870667},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.2434999942779541},{"id":"https://openalex.org/C116834253","wikidata":"https://www.wikidata.org/wiki/Q2039217","display_name":"Identification (biology)","level":2,"score":0.23970000445842743},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.23190000653266907}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.18653/v1/2026.loreslm-1.40","doi":"https://doi.org/10.18653/v1/2026.loreslm-1.40","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.loreslm-1.40","pdf_url":"https://aclanthology.org/2026.loreslm-1.40.pdf","source":{"id":"https://openalex.org/S7407107604","display_name":"Workshop on Language Models for Low-Resource Languages (LoResLM)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"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":"conference"},"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 Second Workshop on Language Models for Low-Resource Languages (LoResLM 2026)","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.18653/v1/2026.loreslm-1.40","doi":"https://doi.org/10.18653/v1/2026.loreslm-1.40","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.loreslm-1.40","pdf_url":"https://aclanthology.org/2026.loreslm-1.40.pdf","source":{"id":"https://openalex.org/S7407107604","display_name":"Workshop on Language Models for Low-Resource Languages (LoResLM)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"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":"conference"},"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 Second Workshop on Language Models for Low-Resource Languages (LoResLM 2026)","raw_type":"proceedings-article"},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7140073660.pdf","grobid_xml":"https://content.openalex.org/works/W7140073660.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W4416429971","https://openalex.org/W3088130516","https://openalex.org/W3088296309","https://openalex.org/W7206160796","https://openalex.org/W7160042364","https://openalex.org/W4416222991","https://openalex.org/W7057397079","https://openalex.org/W3167925372","https://openalex.org/W7131632159","https://openalex.org/W6948340691"],"abstract_inverted_index":{"Encoder-only":[0],"transformers":[1],"remain":[2,22],"essential":[3],"for":[4],"practical":[5,114],"NLP":[6],"tasks.While":[7],"recent":[8,79],"advances":[9],"in":[10,24,116],"multilingual":[11,74,96],"models":[12,67,105],"have":[13],"improved":[14],"crosslingual":[15],"capabilities,":[16],"low-resource":[17],"languages":[18],"such":[19],"as":[20],"Latvian":[21,30,62,117],"underrepresented":[23],"pretraining":[25,38],"corpora,":[26],"and":[27,48,54,64,73,81,98,106,113],"few":[28],"monolingual":[29,72],"encoders":[31,43,75],"currently":[32],"exist.We":[33],"address":[34],"this":[35],"gap":[36],"by":[37],"a":[39,58],"suite":[40],"of":[41,61],"Latvian-specific":[42,100],"based":[44],"on":[45],"RoBERTa,":[46],"DeBERTaV3,":[47],"ModernBERT":[49],"architectures,":[50],"including":[51],"longcontext":[52],"variants,":[53],"evaluating":[55],"them":[56],"across":[57],"diverse":[59],"set":[60],"diagnostic":[63],"linguistic":[65],"benchmarks.Our":[66],"are":[68],"competitive":[69],"with":[70],"existing":[71],"while":[76],"benefiting":[77],"from":[78],"architectural":[80],"efficiency":[82],"advances.Our":[83],"best":[84],"model,":[85],"lv-deberta-base":[86],"(111M":[87],"parameters),":[88],"achieves":[89],"the":[90],"strongest":[91],"overall":[92],"performance,":[93],"outperforming":[94],"larger":[95],"baselines":[97],"prior":[99],"encoders.We":[101],"release":[102],"all":[103],"pretrained":[104],"evaluation":[107],"resources":[108],"to":[109],"support":[110],"further":[111],"research":[112],"applications":[115],"NLP.":[118]},"counts_by_year":[],"updated_date":"2026-10-28T10:33:24.094632","created_date":"2026-03-24T00:00:00"}