{"id":"https://openalex.org/W7134836854","doi":"https://doi.org/10.48550/arxiv.2603.07166","title":"ACD-U: Asymmetric co-teaching with machine unlearning for robust learning with noisy labels","display_name":"ACD-U: Asymmetric co-teaching with machine unlearning for robust learning with noisy labels","publication_year":2026,"publication_date":"2026-03-07","ids":{"openalex":"https://openalex.org/W7134836854","doi":"https://doi.org/10.48550/arxiv.2603.07166"},"language":null,"primary_location":{"id":"pmh:doi:10.48550/arxiv.2603.07166","is_oa":true,"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":"Article"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":null,"any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5115722281","display_name":"Reo Fukunaga","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Fukunaga, Reo","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5031002717","display_name":"Soh Yoshida","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yoshida, Soh","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5089184180","display_name":"Mitsuji Muneyasu","orcid":"https://orcid.org/0000-0002-4492-5991"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Muneyasu, Mitsuji","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"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":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":0.8051999807357788,"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":0.8051999807357788,"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":0.07689999788999557,"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":0.03189999982714653,"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/transformer","display_name":"Transformer","score":0.47360000014305115},{"id":"https://openalex.org/keywords/exploit","display_name":"Exploit","score":0.4512999951839447},{"id":"https://openalex.org/keywords/memorization","display_name":"Memorization","score":0.44369998574256897},{"id":"https://openalex.org/keywords/convolutional-neural-network","display_name":"Convolutional neural network","score":0.4293999969959259},{"id":"https://openalex.org/keywords/deep-learning","display_name":"Deep learning","score":0.4275999963283539},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.40389999747276306},{"id":"https://openalex.org/keywords/deep-neural-networks","display_name":"Deep neural networks","score":0.34630000591278076},{"id":"https://openalex.org/keywords/consistency","display_name":"Consistency (knowledge bases)","score":0.3400000035762787}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8172000050544739},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.7419000267982483},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.520799994468689},{"id":"https://openalex.org/C66322947","wikidata":"https://www.wikidata.org/wiki/Q11658","display_name":"Transformer","level":3,"score":0.47360000014305115},{"id":"https://openalex.org/C165696696","wikidata":"https://www.wikidata.org/wiki/Q11287","display_name":"Exploit","level":2,"score":0.4512999951839447},{"id":"https://openalex.org/C30038468","wikidata":"https://www.wikidata.org/wiki/Q4354775","display_name":"Memorization","level":2,"score":0.44369998574256897},{"id":"https://openalex.org/C81363708","wikidata":"https://www.wikidata.org/wiki/Q17084460","display_name":"Convolutional neural network","level":2,"score":0.4293999969959259},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.4275999963283539},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.40389999747276306},{"id":"https://openalex.org/C2984842247","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep neural networks","level":3,"score":0.34630000591278076},{"id":"https://openalex.org/C2776436953","wikidata":"https://www.wikidata.org/wiki/Q5163215","display_name":"Consistency (knowledge bases)","level":2,"score":0.3400000035762787},{"id":"https://openalex.org/C103088060","wikidata":"https://www.wikidata.org/wiki/Q1062839","display_name":"Error detection and correction","level":2,"score":0.33180001378059387},{"id":"https://openalex.org/C97931131","wikidata":"https://www.wikidata.org/wiki/Q5282087","display_name":"Discriminative model","level":2,"score":0.3314000070095062},{"id":"https://openalex.org/C2776760102","wikidata":"https://www.wikidata.org/wiki/Q5139990","display_name":"Code (set theory)","level":3,"score":0.33079999685287476},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.3050000071525574},{"id":"https://openalex.org/C93361087","wikidata":"https://www.wikidata.org/wiki/Q4426698","display_name":"Data consistency","level":2,"score":0.30169999599456787},{"id":"https://openalex.org/C81917197","wikidata":"https://www.wikidata.org/wiki/Q628760","display_name":"Selection (genetic algorithm)","level":2,"score":0.29989999532699585},{"id":"https://openalex.org/C198531522","wikidata":"https://www.wikidata.org/wiki/Q485146","display_name":"Sample (material)","level":2,"score":0.28700000047683716},{"id":"https://openalex.org/C63479239","wikidata":"https://www.wikidata.org/wiki/Q7353546","display_name":"Robustness (evolution)","level":3,"score":0.2736000120639801},{"id":"https://openalex.org/C148483581","wikidata":"https://www.wikidata.org/wiki/Q446488","display_name":"Feature selection","level":2,"score":0.26499998569488525}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:doi:10.48550/arxiv.2603.07166","is_oa":true,"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":"Article"},{"id":"doi:10.48550/arxiv.2603.07166","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.07166","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":null,"license_id":null,"version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"article"}],"best_oa_location":{"id":"pmh:doi:10.48550/arxiv.2603.07166","is_oa":true,"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":"Article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Deep":[0],"neural":[1,99],"networks":[2,33],"are":[3],"prone":[4],"to":[5,26,181],"memorizing":[6],"incorrect":[7],"labels":[8],"during":[9],"training,":[10],"which":[11],"degrades":[12],"their":[13,103,165],"generalizability.":[14],"Although":[15],"recent":[16],"methods":[17],"have":[18],"combined":[19],"sample":[20,49],"selection":[21,45],"with":[22,59,96],"semi-supervised":[23],"learning":[24,105,175],"(SSL)":[25],"exploit":[27],"the":[28,107,114,122,132,174],"memorization":[29],"effect":[30],"--":[31,41],"where":[32,121],"learn":[34],"from":[35,177],"clean":[36,129],"data":[37,40],"before":[38],"noisy":[39,190],"they":[42],"cannot":[43],"correct":[44],"errors":[46],"once":[47],"a":[48,89,97],"is":[50,125,134,213],"misclassified.":[51],"To":[52],"overcome":[53],"this,":[54],"we":[55],"propose":[56],"asymmetric":[57,64,86],"co-teaching":[58,65,87],"different":[60,69],"architectures":[61,71],"(ACD)-U,":[62],"an":[63],"framework":[66],"that":[67],"uses":[68],"model":[70,109],"and":[72,131,158,162,188,197,207],"incorporates":[73],"machine":[74],"unlearning.":[75],"ACD-U":[76],"addresses":[77],"this":[78],"limitation":[79],"through":[80,136,154],"two":[81],"core":[82],"mechanisms.":[83],"First,":[84],"its":[85],"pairs":[88],"contrastive":[90],"language-image":[91],"pretraining":[92],"(CLIP)-pretrained":[93],"vision":[94,123],"Transformer":[95,124],"convolutional":[98],"network":[100],"(CNN),":[101],"leveraging":[102],"complementary":[104],"behaviors:":[106],"pretrained":[108],"provides":[110],"stable":[111],"predictions,":[112],"whereas":[113],"CNN":[115,133],"adapts":[116],"throughout":[117],"training.":[118],"This":[119,171],"asymmetry,":[120],"trained":[126,135],"only":[127],"on":[128,186],"samples":[130,153],"SSL,":[137],"effectively":[138],"mitigates":[139],"confirmation":[140],"bias.":[141],"Second,":[142],"selective":[143],"unlearning":[144],"enables":[145],"post-hoc":[146],"error":[147,179,183],"correction":[148],"by":[149],"identifying":[150],"incorrectly":[151],"memorized":[152],"loss":[155],"trajectory":[156],"analysis":[157],"CLIP":[159],"consistency":[160],"checks,":[161],"then":[163],"removing":[164],"influence":[166],"via":[167],"Kullback--Leibler":[168],"divergence-based":[169],"forgetting.":[170],"approach":[172],"shifts":[173],"paradigm":[176],"passive":[178],"avoidance":[180],"active":[182],"correction.":[184],"Experiments":[185],"synthetic":[187],"real-world":[189],"datasets,":[191],"including":[192],"CIFAR-10/100,":[193],"CIFAR-N,":[194],"WebVision,":[195],"Clothing1M,":[196],"Red":[198],"Mini-ImageNet,":[199],"demonstrate":[200],"state-of-the-art":[201],"performance,":[202],"particularly":[203],"in":[204],"high-noise":[205],"regimes":[206],"under":[208],"instance-dependent":[209],"noise.":[210],"The":[211],"code":[212],"publicly":[214],"available":[215],"at":[216],"https://github.com/meruemon/ACD-U.":[217]},"counts_by_year":[],"updated_date":"2026-06-11T09:08:48.828518","created_date":"2026-03-11T00:00:00"}