{"id":"https://openalex.org/W4405716273","doi":"https://doi.org/10.1109/wacv61042.2026.00368","title":"Frequency Is What You Need: Considering Word Frequency When Text Masking Benefits Vision-Language Model Pre-training","display_name":"Frequency Is What You Need: Considering Word Frequency When Text Masking Benefits Vision-Language Model Pre-training","publication_year":2026,"publication_date":"2026-03-06","ids":{"openalex":"https://openalex.org/W4405716273","doi":"https://doi.org/10.1109/wacv61042.2026.00368"},"language":"en","primary_location":{"id":"doi:10.1109/wacv61042.2026.00368","is_oa":false,"landing_page_url":"https://doi.org/10.1109/wacv61042.2026.00368","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2026 IEEE/CVF Winter Conference on Applications of Computer Vision (WACV)","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["arxiv","crossref","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2412.16148","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5020087557","display_name":"Mingliang Liang","orcid":"https://orcid.org/0000-0002-7129-6799"},"institutions":[{"id":"https://openalex.org/I145872427","display_name":"Radboud University Nijmegen","ror":"https://ror.org/016xsfp80","country_code":"NL","type":"education","lineage":["https://openalex.org/I145872427"]},{"id":"https://openalex.org/I2802934949","display_name":"Radboud University Medical Center","ror":"https://ror.org/05wg1m734","country_code":"NL","type":"funder","lineage":["https://openalex.org/I2802934949"]},{"id":"https://openalex.org/I4210109357","display_name":"Radboud Institute for Molecular Life Sciences","ror":"https://ror.org/01yb10j39","country_code":"NL","type":"facility","lineage":["https://openalex.org/I145872427","https://openalex.org/I4210109357"]}],"countries":["NL"],"is_corresponding":false,"raw_author_name":"Mingliang Liang","raw_affiliation_strings":["Radboud University,Institute for Computing and Information Sciences,Nijmegen,Netherland"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Radboud University,Institute for Computing and Information Sciences,Nijmegen,Netherland","institution_ids":["https://openalex.org/I145872427","https://openalex.org/I2802934949","https://openalex.org/I4210109357"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5056272341","display_name":"Martha Larson","orcid":"https://orcid.org/0000-0003-4229-5866"},"institutions":[{"id":"https://openalex.org/I145872427","display_name":"Radboud University Nijmegen","ror":"https://ror.org/016xsfp80","country_code":"NL","type":"education","lineage":["https://openalex.org/I145872427"]},{"id":"https://openalex.org/I2802934949","display_name":"Radboud University Medical Center","ror":"https://ror.org/05wg1m734","country_code":"NL","type":"funder","lineage":["https://openalex.org/I2802934949"]},{"id":"https://openalex.org/I4210109357","display_name":"Radboud Institute for Molecular Life Sciences","ror":"https://ror.org/01yb10j39","country_code":"NL","type":"facility","lineage":["https://openalex.org/I145872427","https://openalex.org/I4210109357"]}],"countries":["NL"],"is_corresponding":false,"raw_author_name":"Martha Larson","raw_affiliation_strings":["Radboud University,Institute for Computing and Information Sciences,Nijmegen,Netherland"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Radboud University,Institute for Computing and Information Sciences,Nijmegen,Netherland","institution_ids":["https://openalex.org/I145872427","https://openalex.org/I2802934949","https://openalex.org/I4210109357"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":3,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"3771","last_page":"3780"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10181","display_name":"Natural Language Processing Techniques","score":0.8557999730110168,"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":0.8557999730110168,"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/T12031","display_name":"Speech and dialogue systems","score":0.821399986743927,"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/masking","display_name":"Masking (illustration)","score":0.7244112491607666},{"id":"https://openalex.org/keywords/word","display_name":"Word (group theory)","score":0.5898681879043579},{"id":"https://openalex.org/keywords/word-lists-by-frequency","display_name":"Word lists by frequency","score":0.5688062310218811},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.5604563355445862},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.46720239520072937},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.3972097635269165},{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.34978169202804565},{"id":"https://openalex.org/keywords/audiology","display_name":"Audiology","score":0.32161790132522583},{"id":"https://openalex.org/keywords/linguistics","display_name":"Linguistics","score":0.2666354179382324},{"id":"https://openalex.org/keywords/medicine","display_name":"Medicine","score":0.12090694904327393},{"id":"https://openalex.org/keywords/art","display_name":"Art","score":0.06681936979293823}],"concepts":[{"id":"https://openalex.org/C2777402240","wikidata":"https://www.wikidata.org/wiki/Q6783436","display_name":"Masking (illustration)","level":2,"score":0.7244112491607666},{"id":"https://openalex.org/C90805587","wikidata":"https://www.wikidata.org/wiki/Q10944557","display_name":"Word (group theory)","level":2,"score":0.5898681879043579},{"id":"https://openalex.org/C175293574","wikidata":"https://www.wikidata.org/wiki/Q697133","display_name":"Word lists by frequency","level":3,"score":0.5688062310218811},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5604563355445862},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.46720239520072937},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3972097635269165},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.34978169202804565},{"id":"https://openalex.org/C548259974","wikidata":"https://www.wikidata.org/wiki/Q569965","display_name":"Audiology","level":1,"score":0.32161790132522583},{"id":"https://openalex.org/C41895202","wikidata":"https://www.wikidata.org/wiki/Q8162","display_name":"Linguistics","level":1,"score":0.2666354179382324},{"id":"https://openalex.org/C71924100","wikidata":"https://www.wikidata.org/wiki/Q11190","display_name":"Medicine","level":0,"score":0.12090694904327393},{"id":"https://openalex.org/C142362112","wikidata":"https://www.wikidata.org/wiki/Q735","display_name":"Art","level":0,"score":0.06681936979293823},{"id":"https://openalex.org/C2777530160","wikidata":"https://www.wikidata.org/wiki/Q41796","display_name":"Sentence","level":2,"score":0.0},{"id":"https://openalex.org/C153349607","wikidata":"https://www.wikidata.org/wiki/Q36649","display_name":"Visual arts","level":1,"score":0.0},{"id":"https://openalex.org/C138885662","wikidata":"https://www.wikidata.org/wiki/Q5891","display_name":"Philosophy","level":0,"score":0.0}],"mesh":[],"locations_count":4,"locations":[{"id":"doi:10.1109/wacv61042.2026.00368","is_oa":false,"landing_page_url":"https://doi.org/10.1109/wacv61042.2026.00368","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2026 IEEE/CVF Winter Conference on Applications of Computer Vision (WACV)","raw_type":"proceedings-article"},{"id":"pmh:oai:arXiv.org:2412.16148","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2412.16148","pdf_url":"https://arxiv.org/pdf/2412.16148","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"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","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"pmh:oai:repository.ubn.ru.nl:2066/331537","is_oa":false,"landing_page_url":"https://hdl.handle.net/2066/331537","pdf_url":null,"source":{"id":"https://openalex.org/S4306401067","display_name":"Radboud Repository (Radboud University)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I145872427","host_organization_name":"Radboud University Nijmegen","host_organization_lineage":["https://openalex.org/I145872427"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"https://openaccess.thecvf.com/content/WACV2026/html/Liang_Frequency_Is_What_You_Need_Considering_Word_Frequency_When_Text_WACV_2026_paper.html","raw_type":"Article in monograph or in proceedings"},{"id":"doi:10.48550/arxiv.2412.16148","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2412.16148","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"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","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2412.16148","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2412.16148","pdf_url":"https://arxiv.org/pdf/2412.16148","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"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","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W3081694532","https://openalex.org/W2319684423","https://openalex.org/W2127709934","https://openalex.org/W1588516692","https://openalex.org/W1970935140","https://openalex.org/W1512024959","https://openalex.org/W2017557587","https://openalex.org/W2320411771","https://openalex.org/W4240310336","https://openalex.org/W2346578824"],"abstract_inverted_index":{"Vision":[0],"Language":[1],"Models":[2],"(VLMs)":[3],"can":[4,12],"be":[5,13],"trained":[6],"more":[7],"efficiently":[8],"if":[9],"training":[10,28,69],"sets":[11],"reduced":[14],"in":[15,67],"size.":[16],"Recent":[17],"work":[18],"has":[19,43],"shown":[20],"the":[21,48,56,64,68,104,117],"benefits":[22],"of":[23,32,58,106,119,147],"masking":[24,38,46,61,94,110,133,137,154],"text":[25,60,153],"during":[26,143],"VLM":[27,157],"using":[29],"a":[30,145,152],"variety":[31],"strategies":[33,62],"(truncation,":[34],"random":[35],"masking,":[36],"block":[37],"and":[39,42,71,111],"syntax":[40,45,109,136],"masking)":[41],"reported":[44],"as":[47],"top":[49],"performer.":[50],"In":[51],"this":[52,74],"paper,":[53],"we":[54],"analyze":[55],"impact":[57,75],"different":[59],"on":[63],"word":[65,99],"frequency":[66],"data,":[70],"show":[72,124],"that":[73,125],"is":[76,161],"connected":[77],"to":[78],"model":[79],"success.":[80],"This":[81],"finding":[82,146],"motivates":[83],"Contrastive":[84],"Language-Image":[85],"Pre-training":[86],"with":[87],"Word":[88],"Frequency":[89],"Masking":[90],"(CLIPF),":[91],"our":[92],"proposed":[93],"approach,":[95],"which":[96],"directly":[97],"leverages":[98],"frequency.":[100],"Extensive":[101],"experiments":[102],"demonstrate":[103],"advantages":[105],"CLIPF":[107],"over":[108],"other":[112,131],"existing":[113,132],"approaches,":[114],"particularly":[115],"when":[116,138],"number":[118],"input":[120],"tokens":[121],"decreases.":[122],"We":[123],"not":[126],"only":[127],"CLIPF,":[128],"but":[129],"also":[130],"strategies,":[134],"outperform":[135],"enough":[139],"epochs":[140],"are":[141],"used":[142],"training,":[144],"practical":[148],"importance":[149],"for":[150,156],"selecting":[151],"method":[155],"training.":[158],"Our":[159],"code":[160],"available":[162],"online.":[163]},"counts_by_year":[],"updated_date":"2026-07-25T09:21:30.201066","created_date":"2025-10-10T00:00:00"}