{"id":"https://openalex.org/W4415344501","doi":"https://doi.org/10.1109/iccv51701.2025.00430","title":"Unlearning the Noisy Correspondence Makes CLIP More Robust","display_name":"Unlearning the Noisy Correspondence Makes CLIP More Robust","publication_year":2025,"publication_date":"2025-10-19","ids":{"openalex":"https://openalex.org/W4415344501","doi":"https://doi.org/10.1109/iccv51701.2025.00430"},"language":"en","primary_location":{"id":"doi:10.1109/iccv51701.2025.00430","is_oa":false,"landing_page_url":"https://doi.org/10.1109/iccv51701.2025.00430","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 IEEE/CVF International Conference on Computer Vision (ICCV)","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["arxiv","crossref"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2507.03434","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5101119626","display_name":"Haochen Han","orcid":"https://orcid.org/0000-0002-0479-8932"},"institutions":[{"id":"https://openalex.org/I4210136793","display_name":"Peng Cheng Laboratory","ror":"https://ror.org/03qdqbt06","country_code":"CN","type":"facility","lineage":["https://openalex.org/I4210136793"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Haochen Han","raw_affiliation_strings":["Peng Cheng Laboratory"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Peng Cheng Laboratory","institution_ids":["https://openalex.org/I4210136793"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5032671650","display_name":"Alex Jinpeng Wang","orcid":null},"institutions":[{"id":"https://openalex.org/I4210136793","display_name":"Peng Cheng Laboratory","ror":"https://ror.org/03qdqbt06","country_code":"CN","type":"facility","lineage":["https://openalex.org/I4210136793"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Alex Jinpeng Wang","raw_affiliation_strings":["Peng Cheng Laboratory"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Peng Cheng Laboratory","institution_ids":["https://openalex.org/I4210136793"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5025706580","display_name":"Peijun Ye","orcid":"https://orcid.org/0000-0001-9987-9016"},"institutions":[{"id":"https://openalex.org/I4210136793","display_name":"Peng Cheng Laboratory","ror":"https://ror.org/03qdqbt06","country_code":"CN","type":"facility","lineage":["https://openalex.org/I4210136793"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Peijun Ye","raw_affiliation_strings":["Peng Cheng Laboratory"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Peng Cheng Laboratory","institution_ids":["https://openalex.org/I4210136793"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5048398271","display_name":"Fangming Liu","orcid":"https://orcid.org/0000-0002-8570-1345"},"institutions":[{"id":"https://openalex.org/I4210136793","display_name":"Peng Cheng Laboratory","ror":"https://ror.org/03qdqbt06","country_code":"CN","type":"facility","lineage":["https://openalex.org/I4210136793"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Fangming Liu","raw_affiliation_strings":["Peng Cheng Laboratory"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Peng Cheng Laboratory","institution_ids":["https://openalex.org/I4210136793"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I4210136793"],"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":"4518","last_page":"4528"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T14094","display_name":"Italy: Economic History and Contemporary Issues","score":0.04919999837875366,"subfield":{"id":"https://openalex.org/subfields/2002","display_name":"Economics and Econometrics"},"field":{"id":"https://openalex.org/fields/20","display_name":"Economics, Econometrics and Finance"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}},"topics":[{"id":"https://openalex.org/T14094","display_name":"Italy: Economic History and Contemporary Issues","score":0.04919999837875366,"subfield":{"id":"https://openalex.org/subfields/2002","display_name":"Economics and Econometrics"},"field":{"id":"https://openalex.org/fields/20","display_name":"Economics, Econometrics and Finance"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}},{"id":"https://openalex.org/T12961","display_name":"Regional Development and Policy","score":0.04740000143647194,"subfield":{"id":"https://openalex.org/subfields/3320","display_name":"Political Science and International Relations"},"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/T11653","display_name":"Financial Distress and Bankruptcy Prediction","score":0.0430000014603138,"subfield":{"id":"https://openalex.org/subfields/1402","display_name":"Accounting"},"field":{"id":"https://openalex.org/fields/14","display_name":"Business, Management and Accounting"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/robustness","display_name":"Robustness (evolution)","score":0.6884999871253967},{"id":"https://openalex.org/keywords/noisy-data","display_name":"Noisy data","score":0.5324000120162964},{"id":"https://openalex.org/keywords/false-positive-paradox","display_name":"False positive paradox","score":0.522599995136261},{"id":"https://openalex.org/keywords/process","display_name":"Process (computing)","score":0.4918000102043152},{"id":"https://openalex.org/keywords/forgetting","display_name":"Forgetting","score":0.4860000014305115},{"id":"https://openalex.org/keywords/noise","display_name":"Noise (video)","score":0.43950000405311584},{"id":"https://openalex.org/keywords/perspective","display_name":"Perspective (graphical)","score":0.43059998750686646},{"id":"https://openalex.org/keywords/key","display_name":"Key (lock)","score":0.4242999851703644}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.755299985408783},{"id":"https://openalex.org/C63479239","wikidata":"https://www.wikidata.org/wiki/Q7353546","display_name":"Robustness (evolution)","level":3,"score":0.6884999871253967},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5738000273704529},{"id":"https://openalex.org/C2781170535","wikidata":"https://www.wikidata.org/wiki/Q30587856","display_name":"Noisy data","level":2,"score":0.5324000120162964},{"id":"https://openalex.org/C64869954","wikidata":"https://www.wikidata.org/wiki/Q1859747","display_name":"False positive paradox","level":2,"score":0.522599995136261},{"id":"https://openalex.org/C98045186","wikidata":"https://www.wikidata.org/wiki/Q205663","display_name":"Process (computing)","level":2,"score":0.4918000102043152},{"id":"https://openalex.org/C7149132","wikidata":"https://www.wikidata.org/wiki/Q1377840","display_name":"Forgetting","level":2,"score":0.4860000014305115},{"id":"https://openalex.org/C99498987","wikidata":"https://www.wikidata.org/wiki/Q2210247","display_name":"Noise (video)","level":3,"score":0.43950000405311584},{"id":"https://openalex.org/C12713177","wikidata":"https://www.wikidata.org/wiki/Q1900281","display_name":"Perspective (graphical)","level":2,"score":0.43059998750686646},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.4242999851703644},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.40779998898506165},{"id":"https://openalex.org/C177148314","wikidata":"https://www.wikidata.org/wiki/Q170084","display_name":"Generalization","level":2,"score":0.3427000045776367},{"id":"https://openalex.org/C51632099","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Training set","level":2,"score":0.33219999074935913},{"id":"https://openalex.org/C2776036281","wikidata":"https://www.wikidata.org/wiki/Q48769818","display_name":"Constraint (computer-aided design)","level":2,"score":0.3174000084400177},{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"score":0.31310001015663147},{"id":"https://openalex.org/C24756922","wikidata":"https://www.wikidata.org/wiki/Q1757694","display_name":"Data quality","level":3,"score":0.3077999949455261},{"id":"https://openalex.org/C67186912","wikidata":"https://www.wikidata.org/wiki/Q367664","display_name":"Data modeling","level":2,"score":0.3028999865055084},{"id":"https://openalex.org/C153083717","wikidata":"https://www.wikidata.org/wiki/Q6535263","display_name":"Leverage (statistics)","level":2,"score":0.287200003862381},{"id":"https://openalex.org/C132964779","wikidata":"https://www.wikidata.org/wiki/Q2110223","display_name":"Raw data","level":2,"score":0.28529998660087585},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.2797999978065491},{"id":"https://openalex.org/C160920958","wikidata":"https://www.wikidata.org/wiki/Q7662746","display_name":"Synthetic data","level":2,"score":0.2791000008583069},{"id":"https://openalex.org/C43521106","wikidata":"https://www.wikidata.org/wiki/Q2165493","display_name":"Pipeline (software)","level":2,"score":0.2711000144481659},{"id":"https://openalex.org/C2776760102","wikidata":"https://www.wikidata.org/wiki/Q5139990","display_name":"Code (set theory)","level":3,"score":0.26829999685287476},{"id":"https://openalex.org/C2779530757","wikidata":"https://www.wikidata.org/wiki/Q1207505","display_name":"Quality (philosophy)","level":2,"score":0.25380000472068787}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1109/iccv51701.2025.00430","is_oa":false,"landing_page_url":"https://doi.org/10.1109/iccv51701.2025.00430","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 IEEE/CVF International Conference on Computer Vision (ICCV)","raw_type":"proceedings-article"},{"id":"pmh:oai:arXiv.org:2507.03434","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2507.03434","pdf_url":"https://arxiv.org/pdf/2507.03434","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":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2507.03434","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2507.03434","pdf_url":"https://arxiv.org/pdf/2507.03434","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":null,"license_id":null,"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":[],"abstract_inverted_index":{"The":[0,115,183],"data":[1,24,37,71],"appetite":[2],"for":[3,54,131,152],"Vision-Language":[4],"Models":[5],"(VLMs)":[6],"has":[7],"continuously":[8],"scaled":[9],"up":[10],"from":[11,65],"the":[12,40,87,121,160,171],"early":[13],"millions":[14],"to":[15,68,84,117],"billions":[16],"today,":[17],"which":[18,125],"faces":[19],"an":[20],"untenable":[21],"trade-off":[22],"with":[23,159,179],"quality":[25],"and":[26,135],"inevitably":[27],"introduces":[28],"Noisy":[29,100],"Correspondence":[30,101],"(NC)":[31],"samples.":[32],"Undoubtedly,":[33],"such":[34,59],"semantically":[35],"unrelated":[36],"significantly":[38],"impairs":[39],"performance":[41],"of":[42,90],"VLMs.":[43,94],"Previous":[44],"efforts":[45],"mainly":[46],"address":[47],"this":[48,74],"challenge":[49],"by":[50,110],"estimating":[51],"refined":[52],"alignment":[53],"more":[55],"precise":[56],"guidance.":[57],"However,":[58],"resource-intensive":[60],"pipelines":[61],"that":[62,82,105],"train":[63],"VLMs":[64],"scratch":[66],"struggle":[67],"meet":[69],"realistic":[70],"demands.":[72],"In":[73],"paper,":[75],"we":[76,96],"present":[77],"a":[78,99],"brand":[79],"new":[80],"perspective":[81],"seeks":[83],"directly":[85],"eliminate":[86],"harmful":[88],"effects":[89],"NC":[91],"in":[92],"pre-trained":[93,173],"Specifically,":[95],"propose":[97],"NCU,":[98],"Unlearning":[102],"fine-tuning":[103],"framework":[104],"efficiently":[106],"enhances":[107],"VLMs'":[108],"robustness":[109],"forgetting":[111],"learned":[112],"noisy":[113],"knowledge.":[114],"key":[116],"NCU":[118,169],"is":[119],"learning":[120],"hardest":[122],"negative":[123],"information,":[124],"can":[126,143],"provide":[127],"explicit":[128],"unlearning":[129,141],"direction":[130],"both":[132],"false":[133,136],"positives":[134],"negatives.":[137],"Such":[138],"twin":[139],"goals":[140],"process":[142],"be":[144,186],"formalized":[145],"into":[146],"one":[147],"unified":[148],"optimal":[149],"transport":[150],"objective":[151],"fast":[153],"fine-tuning.":[154],"We":[155],"validate":[156],"our":[157],"approach":[158],"prevailing":[161],"CLIP":[162],"model":[163],"over":[164],"various":[165],"downstream":[166],"tasks.":[167],"Remarkably,":[168],"surpasses":[170],"robust":[172],"method":[174],"on":[175],"zero-shot":[176],"transfer":[177],"while":[178],"lower":[180],"computational":[181],"overhead.":[182],"code":[184],"will":[185],"released":[187],"upon":[188],"acceptance.":[189]},"counts_by_year":[],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-10-20T00:00:00"}