{"id":"https://openalex.org/W4412889581","doi":"https://doi.org/10.18653/v1/2025.acl-long.1546","title":"Beyond Text Compression: Evaluating Tokenizers Across Scales","display_name":"Beyond Text Compression: Evaluating Tokenizers Across Scales","publication_year":2025,"publication_date":"2025-01-01","ids":{"openalex":"https://openalex.org/W4412889581","doi":"https://doi.org/10.18653/v1/2025.acl-long.1546"},"language":"en","primary_location":{"id":"doi:10.18653/v1/2025.acl-long.1546","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.acl-long.1546","pdf_url":"https://aclanthology.org/2025.acl-long.1546.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 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://aclanthology.org/2025.acl-long.1546.pdf","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5081588948","display_name":"Jonas F. Lotz","orcid":"https://orcid.org/0000-0001-6405-0590"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jonas F. Lotz","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5029838043","display_name":"Ant\u00f3nio V. Lopes","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ant\u00f3nio V. Lopes","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5030708442","display_name":"Stephan Peitz","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Stephan Peitz","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101717683","display_name":"Hendra Setiawan","orcid":"https://orcid.org/0000-0003-3071-1383"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Hendra Setiawan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5072554002","display_name":"L Emili","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Leonardo Emili","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":5,"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":0.08689584,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"32155","last_page":"32173"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10181","display_name":"Natural Language Processing Techniques","score":0.7656000256538391,"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":0.7656000256538391,"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/T11902","display_name":"Intelligent Tutoring Systems and Adaptive Learning","score":0.6797000169754028,"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/T13629","display_name":"Text Readability and Simplification","score":0.6708999872207642,"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/computer-science","display_name":"Computer science","score":0.6419021487236023},{"id":"https://openalex.org/keywords/compression","display_name":"Compression (physics)","score":0.5023045539855957},{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.33322665095329285},{"id":"https://openalex.org/keywords/data-science","display_name":"Data science","score":0.33216041326522827},{"id":"https://openalex.org/keywords/physics","display_name":"Physics","score":0.08587807416915894}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6419021487236023},{"id":"https://openalex.org/C180016635","wikidata":"https://www.wikidata.org/wiki/Q2712821","display_name":"Compression (physics)","level":2,"score":0.5023045539855957},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.33322665095329285},{"id":"https://openalex.org/C2522767166","wikidata":"https://www.wikidata.org/wiki/Q2374463","display_name":"Data science","level":1,"score":0.33216041326522827},{"id":"https://openalex.org/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","level":0,"score":0.08587807416915894},{"id":"https://openalex.org/C97355855","wikidata":"https://www.wikidata.org/wiki/Q11473","display_name":"Thermodynamics","level":1,"score":0.0}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.18653/v1/2025.acl-long.1546","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.acl-long.1546","pdf_url":"https://aclanthology.org/2025.acl-long.1546.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 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)","raw_type":"proceedings-article"},{"id":"pmh:oai:pure.atira.dk:openaire_cris_publications/0bca46ec-e417-4690-be86-0e7cd7801d46","is_oa":true,"landing_page_url":"https://researchprofiles.ku.dk/da/publications/0bca46ec-e417-4690-be86-0e7cd7801d46","pdf_url":"https://curis.ku.dk/ws/files/535488852/Beyond_Text_Compression.pdf","source":{"id":"https://openalex.org/S4306401983","display_name":"Research at the University of Copenhagen (University of Copenhagen)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I124055696","host_organization_name":"University of Copenhagen","host_organization_lineage":["https://openalex.org/I124055696"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"Lotz , J F , Lopes , A V , Peitz , S , Setiawan , H & Emili , L 2025 , Beyond Text Compression : Evaluating Tokenizers Across Scales . in W Che , J Nabende , E Shutova & M T Pilehvar (eds) , Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) . Association for Computational Linguistics (ACL) , pp. 32155-32173 , 63rd Annual Meeting of the Association for Computational Linguistics, ACL 2025 , Vienna , Austria , 27/07/2025 . https://doi.org/10.18653/v1/2025.acl-long.1546","raw_type":"contributionToPeriodical"}],"best_oa_location":{"id":"doi:10.18653/v1/2025.acl-long.1546","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.acl-long.1546","pdf_url":"https://aclanthology.org/2025.acl-long.1546.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 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[],"funders":[{"id":"https://openalex.org/F4320327343","display_name":"Rockwool Fonden","ror":"https://ror.org/00c1h4g48"}],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4412889581.pdf","grobid_xml":"https://content.openalex.org/works/W4412889581.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W4391375266","https://openalex.org/W2899084033","https://openalex.org/W2748952813","https://openalex.org/W2390279801","https://openalex.org/W4391913857","https://openalex.org/W2358668433","https://openalex.org/W4396701345","https://openalex.org/W2376932109","https://openalex.org/W2001405890","https://openalex.org/W4396696052"],"abstract_inverted_index":{"The":[0],"choice":[1,59],"of":[2,15,44,106],"tokenizer":[3,16,36,58,79,107,116,126],"can":[4,30],"profoundly":[5],"impact":[6,37],"language":[7,130],"model":[8,131],"performance,":[9],"yet":[10],"accessible":[11],"and":[12,52],"reliable":[13,112],"evaluations":[14],"quality":[17],"remain":[18],"an":[19],"open":[20],"challenge.Inspired":[21],"by":[22,82],"scaling":[23],"consistency,":[24],"we":[25,55,109],"show":[26],"that":[27,57,85],"smaller":[28],"models":[29,40],"accurately":[31],"predict":[32],"significant":[33],"differences":[34,72],"in":[35,65,69,73,128],"on":[38,63],"larger":[39],"at":[41],"a":[42,111,120],"fraction":[43],"the":[45],"compute":[46],"cost.By":[47],"systematically":[48],"evaluating":[49],"both":[50],"English-centric":[51],"multilingual":[53,74],"tokenizers,":[54],"find":[56],"has":[60],"negligible":[61],"effects":[62],"tasks":[64],"English":[66],"but":[67],"results":[68],"consistent":[70],"performance":[71,91],"settings.We":[75],"propose":[76],"new":[77],"intrinsic":[78,115],"metrics":[80,101],"inspired":[81],"Zipf's":[83],"law":[84],"correlate":[86],"more":[87,121],"strongly":[88],"with":[89],"downstream":[90],"than":[92],"text":[93],"compression":[94],"when":[95],"modeling":[96],"unseen":[97],"languages.By":[98],"combining":[99],"several":[100],"to":[102,124],"capture":[103],"multiple":[104],"aspects":[105],"behavior,":[108],"develop":[110],"framework":[113],"for":[114],"evaluations.Our":[117],"work":[118],"offers":[119],"efficient":[122],"path":[123],"informed":[125],"selection":[127],"future":[129],"development.":[132]},"counts_by_year":[],"updated_date":"2026-06-11T09:08:48.828518","created_date":"2025-10-10T00:00:00"}