{"id":"https://openalex.org/W7133309084","doi":"https://doi.org/10.48550/arxiv.2603.00404","title":"USE: Uncertainty Structure Estimation for Robust Semi-Supervised Learning","display_name":"USE: Uncertainty Structure Estimation for Robust Semi-Supervised Learning","publication_year":2026,"publication_date":"2026-02-28","ids":{"openalex":"https://openalex.org/W7133309084","doi":"https://doi.org/10.48550/arxiv.2603.00404"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2603.00404","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.00404","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":false,"raw_source_name":null,"raw_type":"article"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2603.00404","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5127921651","display_name":"Tsao-Lun Chen","orcid":null},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Chen, Tsao-Lun","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5127887119","display_name":"Chien-Liang Liu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Liu, Chien-Liang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5010641101","display_name":"Tzu-Ming Harry Hsu","orcid":"https://orcid.org/0000-0001-7198-7832"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Hsu, Tzu-Ming Harry","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5127968062","display_name":"Tai-Hsien Wu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wu, Tai-Hsien","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5065684335","display_name":"Chi-Cheng Fu","orcid":"https://orcid.org/0000-0001-8846-3808"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Fu, Chi-Cheng","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5127964557","display_name":"Han-Yi E. Chou","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chou, Han-Yi E.","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5127975513","display_name":"Shun-Feng Su","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Su, Shun-Feng","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":7,"corresponding_author_ids":["https://openalex.org/A5127921651"],"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/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.2863999903202057,"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/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.2863999903202057,"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/T12535","display_name":"Machine Learning and Data Classification","score":0.20149999856948853,"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/T11512","display_name":"Anomaly Detection Techniques and Applications","score":0.07119999825954437,"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/robustness","display_name":"Robustness (evolution)","score":0.6395999789237976},{"id":"https://openalex.org/keywords/preprocessor","display_name":"Preprocessor","score":0.5705999732017517},{"id":"https://openalex.org/keywords/bottleneck","display_name":"Bottleneck","score":0.5697000026702881},{"id":"https://openalex.org/keywords/entropy","display_name":"Entropy (arrow of time)","score":0.5333999991416931},{"id":"https://openalex.org/keywords/labeled-data","display_name":"Labeled data","score":0.3783999979496002},{"id":"https://openalex.org/keywords/training-set","display_name":"Training set","score":0.3617999851703644},{"id":"https://openalex.org/keywords/cross-entropy","display_name":"Cross entropy","score":0.3580000102519989},{"id":"https://openalex.org/keywords/estimator","display_name":"Estimator","score":0.3560999929904938},{"id":"https://openalex.org/keywords/reliability","display_name":"Reliability (semiconductor)","score":0.33980000019073486}],"concepts":[{"id":"https://openalex.org/C63479239","wikidata":"https://www.wikidata.org/wiki/Q7353546","display_name":"Robustness (evolution)","level":3,"score":0.6395999789237976},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6363999843597412},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6225000023841858},{"id":"https://openalex.org/C34736171","wikidata":"https://www.wikidata.org/wiki/Q918333","display_name":"Preprocessor","level":2,"score":0.5705999732017517},{"id":"https://openalex.org/C2780513914","wikidata":"https://www.wikidata.org/wiki/Q18210350","display_name":"Bottleneck","level":2,"score":0.5697000026702881},{"id":"https://openalex.org/C106301342","wikidata":"https://www.wikidata.org/wiki/Q4117933","display_name":"Entropy (arrow of time)","level":2,"score":0.5333999991416931},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.5027999877929688},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.446399986743927},{"id":"https://openalex.org/C2776145971","wikidata":"https://www.wikidata.org/wiki/Q30673951","display_name":"Labeled data","level":2,"score":0.3783999979496002},{"id":"https://openalex.org/C51632099","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Training set","level":2,"score":0.3617999851703644},{"id":"https://openalex.org/C167981619","wikidata":"https://www.wikidata.org/wiki/Q1685498","display_name":"Cross entropy","level":3,"score":0.3580000102519989},{"id":"https://openalex.org/C185429906","wikidata":"https://www.wikidata.org/wiki/Q1130160","display_name":"Estimator","level":2,"score":0.3560999929904938},{"id":"https://openalex.org/C43214815","wikidata":"https://www.wikidata.org/wiki/Q7310987","display_name":"Reliability (semiconductor)","level":3,"score":0.33980000019073486},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.3287999927997589},{"id":"https://openalex.org/C24756922","wikidata":"https://www.wikidata.org/wiki/Q1757694","display_name":"Data quality","level":3,"score":0.32429999113082886},{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"score":0.31450000405311584},{"id":"https://openalex.org/C58973888","wikidata":"https://www.wikidata.org/wiki/Q1041418","display_name":"Semi-supervised learning","level":2,"score":0.30970001220703125},{"id":"https://openalex.org/C10551718","wikidata":"https://www.wikidata.org/wiki/Q5227332","display_name":"Data pre-processing","level":2,"score":0.3077999949455261},{"id":"https://openalex.org/C168167062","wikidata":"https://www.wikidata.org/wiki/Q1117970","display_name":"Component (thermodynamics)","level":2,"score":0.30219998955726624},{"id":"https://openalex.org/C60008888","wikidata":"https://www.wikidata.org/wiki/Q6031013","display_name":"Information bottleneck method","level":3,"score":0.2994000017642975},{"id":"https://openalex.org/C2779530757","wikidata":"https://www.wikidata.org/wiki/Q1207505","display_name":"Quality (philosophy)","level":2,"score":0.2973000109195709},{"id":"https://openalex.org/C9357733","wikidata":"https://www.wikidata.org/wiki/Q6878417","display_name":"Missing data","level":2,"score":0.2944999933242798},{"id":"https://openalex.org/C149441793","wikidata":"https://www.wikidata.org/wiki/Q200726","display_name":"Probability distribution","level":2,"score":0.2599000036716461},{"id":"https://openalex.org/C58489278","wikidata":"https://www.wikidata.org/wiki/Q1172284","display_name":"Data set","level":2,"score":0.25929999351501465},{"id":"https://openalex.org/C160920958","wikidata":"https://www.wikidata.org/wiki/Q7662746","display_name":"Synthetic data","level":2,"score":0.2572999894618988},{"id":"https://openalex.org/C9679016","wikidata":"https://www.wikidata.org/wiki/Q1417473","display_name":"Principle of maximum entropy","level":2,"score":0.2538999915122986}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2603.00404","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.00404","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":"doi:10.48550/arxiv.2603.00404","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.00404","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":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":{"In":[0,48],"this":[1],"study,":[2],"a":[3,10,104,121,127,142,197,205],"novel":[4],"idea,":[5],"Uncertainty":[6],"Structure":[7],"Estimation":[8],"(USE),":[9],"lightweight,":[11],"algorithm-agnostic":[12],"procedure":[13],"that":[14,74,130,169,187],"emphasizes":[15],"the":[16,42,45,75,86,95,108,188],"often-overlooked":[17],"role":[18],"of":[19,44,88,97,179],"unlabeled":[20,46,50,98,116,149,192],"data":[21,51,150,193],"quality":[22,43,96,194],"is":[23,39,167],"introduced":[24],"for":[25,115,208],"Semi-supervised":[26],"learning":[27],"(SSL).":[28],"SSL":[29,152,212],"has":[30],"achieved":[31],"impressive":[32],"progress,":[33],"but":[34,83],"its":[35],"reliability":[36],"in":[37,69,80,85,213],"deployment":[38],"limited":[40],"by":[41,56],"pool.":[47],"practice,":[49],"are":[52],"almost":[53],"always":[54],"contaminated":[55],"out-of-distribution":[57],"(OOD)":[58],"samples,":[59,117],"where":[60],"both":[61],"near-OOD":[62],"and":[63,93,118,161,174,201,210],"far-OOD":[64],"can":[65,184],"negatively":[66],"affect":[67],"performance":[68],"different":[70],"ways.":[71],"We":[72],"argue":[73],"bottleneck":[76],"does":[77],"not":[78],"lie":[79],"algorithmic":[81],"design,":[82],"rather":[84],"absence":[87],"principled":[89],"mechanisms":[90],"to":[91,111],"assess":[92],"curate":[94],"data.":[99],"The":[100],"proposed":[101,189],"USE":[102,170],"trains":[103],"proxy":[105],"model":[106],"on":[107,158],"labeled":[109],"set":[110],"compute":[112],"entropy":[113],"scores":[114],"then":[119],"derives":[120],"threshold,":[122],"via":[123],"statistical":[124],"comparison":[125],"against":[126],"reference":[128],"distribution,":[129],"separates":[131],"informative":[132],"(structured)":[133],"from":[134],"uninformative":[135,146],"(structureless)":[136],"samples.":[137],"This":[138],"enables":[139],"assessment":[140,199],"as":[141,196,204],"preprocessing":[143],"step,":[144],"removing":[145],"or":[147],"harmful":[148],"before":[151],"training":[153],"begins.":[154],"Through":[155],"extensive":[156],"experiments":[157],"imaging":[159],"(CIFAR-100)":[160],"NLP":[162],"(Yelp":[163],"Review)":[164],"data,":[165],"it":[166,183,203],"evident":[168],"consistently":[171],"improves":[172],"accuracy":[173],"robustness":[175],"under":[176],"varying":[177],"levels":[178],"OOD":[180],"contamination.":[181],"Thus,":[182],"be":[185],"concluded":[186],"approach":[190],"reframes":[191],"control":[195],"structural":[198],"problem,":[200],"considers":[202],"necessary":[206],"component":[207],"reliable":[209],"efficient":[211],"realistic":[214],"mixed-distribution":[215],"environments.":[216]},"counts_by_year":[],"updated_date":"2026-05-05T08:41:31.759640","created_date":"2026-03-04T00:00:00"}