{"id":"https://openalex.org/W4412888322","doi":"https://doi.org/10.18653/v1/2025.findings-acl.609","title":"Cheap Character Noise for OCR-Robust Multilingual Embeddings","display_name":"Cheap Character Noise for OCR-Robust Multilingual Embeddings","publication_year":2025,"publication_date":"2025-01-01","ids":{"openalex":"https://openalex.org/W4412888322","doi":"https://doi.org/10.18653/v1/2025.findings-acl.609"},"language":"en","primary_location":{"id":"doi:10.18653/v1/2025.findings-acl.609","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-acl.609","pdf_url":"https://aclanthology.org/2025.findings-acl.609.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":"Findings of the Association for Computational Linguistics: ACL 2025","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://aclanthology.org/2025.findings-acl.609.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5008587249","display_name":"Andrianos Michail","orcid":null},"institutions":[{"id":"https://openalex.org/I202697423","display_name":"University of Zurich","ror":"https://ror.org/02crff812","country_code":"CH","type":"education","lineage":["https://openalex.org/I202697423"]}],"countries":["CH"],"is_corresponding":true,"raw_author_name":"Andrianos Michail","raw_affiliation_strings":["University of Zurich"],"affiliations":[{"raw_affiliation_string":"University of Zurich","institution_ids":["https://openalex.org/I202697423"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5010534476","display_name":"Juri Opitz","orcid":null},"institutions":[{"id":"https://openalex.org/I202697423","display_name":"University of Zurich","ror":"https://ror.org/02crff812","country_code":"CH","type":"education","lineage":["https://openalex.org/I202697423"]}],"countries":["CH"],"is_corresponding":false,"raw_author_name":"Juri Opitz","raw_affiliation_strings":["University of Zurich"],"affiliations":[{"raw_affiliation_string":"University of Zurich","institution_ids":["https://openalex.org/I202697423"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100424577","display_name":"Yining Wang","orcid":"https://orcid.org/0000-0001-9410-0392"},"institutions":[{"id":"https://openalex.org/I202697423","display_name":"University of Zurich","ror":"https://ror.org/02crff812","country_code":"CH","type":"education","lineage":["https://openalex.org/I202697423"]}],"countries":["CH"],"is_corresponding":false,"raw_author_name":"Yining Wang","raw_affiliation_strings":["University of Zurich"],"affiliations":[{"raw_affiliation_string":"University of Zurich","institution_ids":["https://openalex.org/I202697423"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Robin Meister","orcid":null},"institutions":[{"id":"https://openalex.org/I202697423","display_name":"University of Zurich","ror":"https://ror.org/02crff812","country_code":"CH","type":"education","lineage":["https://openalex.org/I202697423"]}],"countries":["CH"],"is_corresponding":false,"raw_author_name":"Robin Meister","raw_affiliation_strings":["University of Zurich"],"affiliations":[{"raw_affiliation_string":"University of Zurich","institution_ids":["https://openalex.org/I202697423"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5005771535","display_name":"Rico Sennrich","orcid":"https://orcid.org/0000-0002-1438-4741"},"institutions":[{"id":"https://openalex.org/I202697423","display_name":"University of Zurich","ror":"https://ror.org/02crff812","country_code":"CH","type":"education","lineage":["https://openalex.org/I202697423"]}],"countries":["CH"],"is_corresponding":false,"raw_author_name":"Rico Sennrich","raw_affiliation_strings":["University of Zurich"],"affiliations":[{"raw_affiliation_string":"University of Zurich","institution_ids":["https://openalex.org/I202697423"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5073027507","display_name":"Simon Clematide","orcid":"https://orcid.org/0000-0003-1365-0662"},"institutions":[{"id":"https://openalex.org/I202697423","display_name":"University of Zurich","ror":"https://ror.org/02crff812","country_code":"CH","type":"education","lineage":["https://openalex.org/I202697423"]}],"countries":["CH"],"is_corresponding":false,"raw_author_name":"Simon Clematide","raw_affiliation_strings":["University of Zurich"],"affiliations":[{"raw_affiliation_string":"University of Zurich","institution_ids":["https://openalex.org/I202697423"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":6,"corresponding_author_ids":["https://openalex.org/A5008587249"],"corresponding_institution_ids":["https://openalex.org/I202697423"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":true,"cited_by_count":0,"citation_normalized_percentile":{"value":0.09285216,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"11705","last_page":"11716"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10181","display_name":"Natural Language Processing Techniques","score":0.9979000091552734,"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":0.9979000091552734,"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/T10201","display_name":"Speech Recognition and Synthesis","score":0.9940999746322632,"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/T10028","display_name":"Topic Modeling","score":0.9642000198364258,"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/character","display_name":"Character (mathematics)","score":0.7462258338928223},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.6888744831085205},{"id":"https://openalex.org/keywords/optical-character-recognition","display_name":"Optical character recognition","score":0.6527884602546692},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.5517527461051941},{"id":"https://openalex.org/keywords/noise","display_name":"Noise (video)","score":0.5468631386756897},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.522857666015625},{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.5228263139724731},{"id":"https://openalex.org/keywords/character-recognition","display_name":"Character recognition","score":0.49277809262275696},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.1292521059513092},{"id":"https://openalex.org/keywords/image","display_name":"Image (mathematics)","score":0.09750238060951233}],"concepts":[{"id":"https://openalex.org/C2780861071","wikidata":"https://www.wikidata.org/wiki/Q1062934","display_name":"Character (mathematics)","level":2,"score":0.7462258338928223},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6888744831085205},{"id":"https://openalex.org/C546480517","wikidata":"https://www.wikidata.org/wiki/Q167555","display_name":"Optical character recognition","level":3,"score":0.6527884602546692},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.5517527461051941},{"id":"https://openalex.org/C99498987","wikidata":"https://www.wikidata.org/wiki/Q2210247","display_name":"Noise (video)","level":3,"score":0.5468631386756897},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.522857666015625},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.5228263139724731},{"id":"https://openalex.org/C2987247673","wikidata":"https://www.wikidata.org/wiki/Q167555","display_name":"Character recognition","level":3,"score":0.49277809262275696},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.1292521059513092},{"id":"https://openalex.org/C115961682","wikidata":"https://www.wikidata.org/wiki/Q860623","display_name":"Image (mathematics)","level":2,"score":0.09750238060951233},{"id":"https://openalex.org/C2524010","wikidata":"https://www.wikidata.org/wiki/Q8087","display_name":"Geometry","level":1,"score":0.0}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.18653/v1/2025.findings-acl.609","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-acl.609","pdf_url":"https://aclanthology.org/2025.findings-acl.609.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":"Findings of the Association for Computational Linguistics: ACL 2025","raw_type":"proceedings-article"},{"id":"pmh:doi:10.5167/uzh-292769","is_oa":false,"landing_page_url":null,"pdf_url":null,"source":{"id":"https://openalex.org/S4406922384","display_name":"Open MIND","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":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"ConferenceItem"}],"best_oa_location":{"id":"doi:10.18653/v1/2025.findings-acl.609","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-acl.609","pdf_url":"https://aclanthology.org/2025.findings-acl.609.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":"Findings of the Association for Computational Linguistics: ACL 2025","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G5007414595","display_name":"Impresso - Media Monitoring of the Past II. Beyond Borders: Connecting Historical Newspapers and Radio.","funder_award_id":"213585","funder_id":"https://openalex.org/F4320320924","funder_display_name":"Schweizerischer Nationalfonds zur F\u00f6rderung der Wissenschaftlichen Forschung"}],"funders":[{"id":"https://openalex.org/F4320306076","display_name":"National Science Foundation","ror":"https://ror.org/021nxhr62"},{"id":"https://openalex.org/F4320320924","display_name":"Schweizerischer Nationalfonds zur F\u00f6rderung der Wissenschaftlichen Forschung","ror":"https://ror.org/00yjd3n13"},{"id":"https://openalex.org/F4320321038","display_name":"Fonds National de la Recherche Luxembourg","ror":"https://ror.org/039z13y21"}],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4412888322.pdf","grobid_xml":"https://content.openalex.org/works/W4412888322.grobid-xml"},"referenced_works_count":1,"referenced_works":["https://openalex.org/W4401201934"],"related_works":["https://openalex.org/W1991513203","https://openalex.org/W2892704491","https://openalex.org/W3178467699","https://openalex.org/W4312612713","https://openalex.org/W4408696804","https://openalex.org/W2106936331","https://openalex.org/W1985626581","https://openalex.org/W2042634493","https://openalex.org/W2100798736","https://openalex.org/W2728450881"],"abstract_inverted_index":{"The":[0],"large":[1],"amount":[2],"of":[3,28,45,164],"text":[4],"collections":[5,21],"digitized":[6,127,156],"by":[7],"imperfect":[8],"OCR":[9,29,169],"systems":[10],"requires":[11],"semantic":[12,123],"search":[13,124],"models":[14,82,93,116,135,145],"that":[15,38,58,70,91,114,142],"perform":[16],"robustly":[17],"on":[18,108],"noisy":[19,84],"input.Such":[20],"are":[22,39],"highly":[23],"heterogeneous,":[24,126],"with":[25,49,64],"varying":[26],"degrees":[27],"quality,":[30],"spelling":[31],"conventions":[32],"and":[33,67,86,104,134],"other":[34,176],"inconsistencies":[35],"-all":[36],"phenomena":[37],"underrepresented":[40],"in":[41,125,155,181],"the":[42],"training":[43],"data":[44],"standard":[46,109],"embedding":[47,115,144],"models,":[48],"ramifications":[50],"for":[51,121],"their":[52],"generalization.In":[53],"our":[54,131],"paper,":[55],"we":[56,78],"show":[57,90,94],"this":[59],"problem":[60],"can":[61,117,146,173],"be":[62,118,147],"alleviated":[63],"a":[65,87],"simple":[66],"inexpensive":[68],"method":[69],"does":[71],"not":[72,161],"require":[73],"supervision":[74],"or":[75],"in-domain":[76],"training.Specifically,":[77],"fine-tune":[79],"existing":[80],"multilingual":[81,143],"using":[83],"texts":[85],"contrastive":[88],"loss.We":[89],"these":[92],"considerable":[95],"improvements":[96],"across":[97],"different":[98],"noise":[99,152,159],"conditions.Control":[100],"experiments":[101],"indicate":[102],"minimal,":[103],"occasionally":[105],"positive,":[106],"impact":[107],"similarity":[110],"tasks.These":[111],"findings":[112],"suggest":[113],"inexpensively":[119],"adapted":[120],"cross-lingual":[122],"corpora.We":[128],"publicly":[129],"release":[130],"code,":[132],"datasets,":[133],"at":[136],"https://github.com/impresso/":[137],"ocr-robust-multilingual-embeddings.":[138],"LimitationsOur":[139],"study":[140],"demonstrates":[141],"made":[148],"more":[149],"robust":[150],"to":[151],"commonly":[153],"found":[154],"texts.However,":[157],"such":[158,178],"may":[160],"only":[162],"consist":[163],"actual":[165],"errors":[166],"(e.g.,":[167],"from":[168],"systems),":[170],"but":[171],"it":[172],"also":[174],"include":[175],"phenomena,":[177],"as":[179],"typos":[180],"newspapers,":[182],"regional":[183]},"counts_by_year":[],"updated_date":"2026-04-21T08:09:41.155169","created_date":"2025-10-10T00:00:00"}