{"id":"https://openalex.org/W4403573545","doi":"https://doi.org/10.48550/arxiv.2410.11215","title":"A CLIP-Powered Framework for Robust and Generalizable Data Selection","display_name":"A CLIP-Powered Framework for Robust and Generalizable Data Selection","publication_year":2024,"publication_date":"2024-10-15","ids":{"openalex":"https://openalex.org/W4403573545","doi":"https://doi.org/10.48550/arxiv.2410.11215"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2410.11215","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2410.11215","pdf_url":"https://arxiv.org/pdf/2410.11215","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"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"},"volume":null,"type":"preprint","is_traditional":true,"indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2410.11215","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":null,"display_name":"Yang, Suorong","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I4391012619","display_name":"Shanghai Artificial Intelligence Laboratory","ror":"https://ror.org/03wkvpx79","country_code":"CN","type":"facility","lineage":["https://openalex.org/I4391012619"]},{"id":"https://openalex.org/I881766915","display_name":"Nanjing University","ror":"https://ror.org/01rxvg760","country_code":"CN","type":"education","lineage":["https://openalex.org/I881766915"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yang, Suorong","raw_affiliation_strings":["National Key Laboratory for Novel Software Technology , Nanjing University","Shanghai Artificial Intelligence Laboratory"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"National Key Laboratory for Novel Software Technology , Nanjing University","institution_ids":["https://openalex.org/I881766915"]},{"raw_affiliation_string":"Shanghai Artificial Intelligence Laboratory","institution_ids":["https://openalex.org/I4391012619"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5114723643","display_name":"Peng Ye","orcid":"https://orcid.org/0000-0002-8928-0912","observed_orcids":["https://orcid.org/0000-0002-8928-0912"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ye, Peng","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5087818121","display_name":"Wanli Ouyang","orcid":"https://orcid.org/0000-0002-9163-2761","observed_orcids":["https://orcid.org/0000-0002-9163-2761"]},"institutions":[{"id":"https://openalex.org/I4391012619","display_name":"Shanghai Artificial Intelligence Laboratory","ror":"https://ror.org/03wkvpx79","country_code":"CN","type":"facility","lineage":["https://openalex.org/I4391012619"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Ouyang, Wanli","raw_affiliation_strings":["Shanghai Artificial Intelligence Laboratory"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Shanghai Artificial Intelligence Laboratory","institution_ids":["https://openalex.org/I4391012619"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5112578541","display_name":"Dongzhan Zhou","orcid":"https://orcid.org/0000-0001-6568-5440","observed_orcids":["https://orcid.org/0000-0001-6568-5440","https://orcid.org/0009-0007-2513-9585"]},"institutions":[{"id":"https://openalex.org/I4391012619","display_name":"Shanghai Artificial Intelligence Laboratory","ror":"https://ror.org/03wkvpx79","country_code":"CN","type":"facility","lineage":["https://openalex.org/I4391012619"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zhou, Dongzhan","raw_affiliation_strings":["Shanghai Artificial Intelligence Laboratory"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Shanghai Artificial Intelligence Laboratory","institution_ids":["https://openalex.org/I4391012619"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5036608458","display_name":"Furao Shen","orcid":"https://orcid.org/0000-0002-7285-326X","observed_orcids":["https://orcid.org/0000-0002-7285-326X","https://orcid.org/0000-0002-0867-7628"]},"institutions":[{"id":"https://openalex.org/I881766915","display_name":"Nanjing University","ror":"https://ror.org/01rxvg760","country_code":"CN","type":"education","lineage":["https://openalex.org/I881766915"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Shen, Furao","raw_affiliation_strings":["National Key Laboratory for Novel Software Technology , Nanjing University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"National Key Laboratory for Novel Software Technology , Nanjing University","institution_ids":["https://openalex.org/I881766915"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"cited_by_count":16,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":null,"last_page":null},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T12535","display_name":"Machine Learning and Data Classification","score":0.9994999766349792,"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/T12535","display_name":"Machine Learning and Data Classification","score":0.9994999766349792,"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/T10036","display_name":"Advanced Neural Network Applications","score":0.00019999999494757503,"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"}},{"id":"https://openalex.org/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":9.999999747378752e-05,"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/dataset-selection","display_name":"data selection","score":0.9729999899864197},{"id":"https://openalex.org/keywords/clip","display_name":"CLIP","score":0.7049999833106995},{"id":"https://openalex.org/keywords/multimodal-representation-learning","display_name":"multimodal representation learning","score":0.7099999785423279},{"id":"https://openalex.org/keywords/noisy-data-removal","display_name":"noisy data removal","score":0.6959999799728394},{"id":"https://openalex.org/keywords/multi-objective-optimization","display_name":"multi-objective optimization","score":0.8299999833106995},{"id":"https://openalex.org/keywords/training-cost-reduction","display_name":"training cost reduction","score":0.8080000281333923},{"id":"https://openalex.org/keywords/data-quality-improvement","display_name":"data quality improvement","score":0.5249999761581421}],"concepts":[{"id":"https://openalex.org/C81917197","wikidata":"https://www.wikidata.org/wiki/Q628760","display_name":"Selection (genetic algorithm)","level":2,"score":0.6278217434883118},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5863621234893799},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.29067009687423706}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2410.11215","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2410.11215","pdf_url":"https://arxiv.org/pdf/2410.11215","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"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"},{"id":"doi:10.48550/arxiv.2410.11215","doi":"https://doi.org/10.48550/arxiv.2410.11215","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2410.11215","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"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":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2410.11215","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2410.11215","pdf_url":"https://arxiv.org/pdf/2410.11215","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"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":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4403573545.pdf","grobid_xml":"https://content.openalex.org/works/W4403573545.grobid-xml"},"referenced_works_count":16,"referenced_works":["https://openalex.org/W1686810756","https://openalex.org/W2242818861","https://openalex.org/W2741430497","https://openalex.org/W2903996579","https://openalex.org/W3094502228","https://openalex.org/W3215230805","https://openalex.org/W4226002507","https://openalex.org/W4297808394","https://openalex.org/W4322718191","https://openalex.org/W4387415233","https://openalex.org/W4387634985","https://openalex.org/W4389650301","https://openalex.org/W4389983309","https://openalex.org/W4392617488","https://openalex.org/W4401929513","https://openalex.org/W7103751069"],"related_works":["https://openalex.org/W4415182286","https://openalex.org/W4402679785","https://openalex.org/W7083204234","https://openalex.org/W7128070082","https://openalex.org/W4415798716","https://openalex.org/W4415309764","https://openalex.org/W4381568221","https://openalex.org/W4415067568","https://openalex.org/W4416010234","https://openalex.org/W4391590700"],"abstract_inverted_index":{"Large-scale":[0],"datasets":[1,20,30],"have":[2],"been":[3],"pivotal":[4],"to":[5,64,80,90,144,186,203],"the":[6,54,59,66,151,182],"advancements":[7],"of":[8,128],"deep":[9],"learning":[10],"models":[11],"in":[12,52],"recent":[13],"years,":[14],"but":[15,206],"training":[16,42,71,205],"on":[17,41,77,168],"such":[18],"large":[19],"invariably":[21],"incurs":[22],"substantial":[23],"storage":[24],"and":[25,34,44,120,135,149],"computational":[26],"overhead.":[27],"Meanwhile,":[28],"real-world":[29],"often":[31],"contain":[32],"redundant":[33],"noisy":[35,97,177],"data,":[36],"imposing":[37],"a":[38,107,201],"negative":[39],"impact":[40],"efficiency":[43],"model":[45],"performance.":[46],"Data":[47],"selection":[48,111,136,152],"has":[49],"shown":[50],"promise":[51],"identifying":[53],"most":[55],"representative":[56],"samples":[57,180],"from":[58,181],"entire":[60],"dataset,":[61,183],"which":[62,87],"aims":[63],"minimize":[65],"performance":[67,190],"gap":[68],"with":[69,96,191],"reduced":[70],"costs.":[72],"Existing":[73],"works":[74],"typically":[75],"rely":[76],"single-modality":[78],"information":[79,116],"assign":[81],"importance":[82],"scores":[83],"for":[84,117],"individual":[85],"samples,":[86],"may":[88],"lead":[89],"inaccurate":[91],"assessments,":[92],"especially":[93],"when":[94],"dealing":[95],"or":[98,178],"corrupted":[99],"samples.":[100],"To":[101],"address":[102],"this":[103],"limitation,":[104],"we":[105],"propose":[106],"novel":[108],"CLIP-powered":[109],"data":[110,211],"framework":[112,126],"that":[113,160,196],"leverages":[114],"multimodal":[115,142],"more":[118],"robust":[119],"generalizable":[121],"sample":[122,133,147],"selection.":[123],"Specifically,":[124],"our":[125,161,173],"consists":[127],"three":[129],"key":[130],"modules-dataset":[131],"adaptation,":[132],"scoring,":[134],"optimization-that":[137],"together":[138],"harness":[139],"extensive":[140],"pre-trained":[141],"knowledge":[143],"comprehensively":[145],"assess":[146],"influence":[148],"optimize":[150],"results":[153],"through":[154],"multi-objective":[155],"optimization.":[156],"Extensive":[157],"experiments":[158],"demonstrate":[159],"approach":[162],"consistently":[163],"outperforms":[164],"existing":[165],"state-of-the-art":[166],"baselines":[167],"various":[169],"benchmark":[170],"datasets.":[171],"Notably,":[172],"method":[174],"effectively":[175],"removes":[176],"damaged":[179],"enabling":[184],"it":[185,197],"achieve":[187],"even":[188],"higher":[189],"less":[192],"data.":[193],"This":[194],"indicates":[195],"is":[198],"not":[199],"only":[200],"way":[202],"accelerate":[204],"can":[207],"also":[208],"improve":[209],"overall":[210],"quality.":[212]},"counts_by_year":[{"year":2026,"cited_by_count":5},{"year":2025,"cited_by_count":11}],"updated_date":"2026-11-07T11:25:07.637180","created_date":"2024-10-20T00:00:00"}