{"id":"https://openalex.org/W7160862941","doi":"https://doi.org/10.1109/access.2026.3692050","title":"Efficient Data Pruning via Early-Stage Prediction Stability and Balanced Sampling","display_name":"Efficient Data Pruning via Early-Stage Prediction Stability and Balanced Sampling","publication_year":2026,"publication_date":"2026-01-01","ids":{"openalex":"https://openalex.org/W7160862941","doi":"https://doi.org/10.1109/access.2026.3692050"},"language":"en","primary_location":{"id":"doi:10.1109/access.2026.3692050","is_oa":true,"landing_page_url":"https://doi.org/10.1109/access.2026.3692050","pdf_url":null,"source":{"id":"https://openalex.org/S2485537415","display_name":"IEEE Access","issn_l":"2169-3536","issn":["2169-3536"],"is_oa":true,"is_in_doaj":true,"is_core":true,"host_organization":"https://openalex.org/P4310319808","host_organization_name":"Institute of Electrical and Electronics Engineers","host_organization_lineage":["https://openalex.org/P4310319808"],"host_organization_lineage_names":["Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Access","raw_type":"journal-article"},"type":"article","indexed_in":["crossref","doaj"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://doi.org/10.1109/access.2026.3692050","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5020358535","display_name":"Chih-Yao Hu","orcid":"https://orcid.org/0009-0008-5968-6322"},"institutions":[{"id":"https://openalex.org/I134161618","display_name":"National Taiwan Normal University","ror":"https://ror.org/059dkdx38","country_code":"TW","type":"education","lineage":["https://openalex.org/I134161618"]}],"countries":["TW"],"is_corresponding":false,"raw_author_name":"Chih-Yao Hu","raw_affiliation_strings":["Department of Electrical Engineering, National Taiwan Normal University, Taipei, Taiwan"],"raw_orcid":"https://orcid.org/0009-0004-4681-2810","affiliations":[{"raw_affiliation_string":"Department of Electrical Engineering, National Taiwan Normal University, Taipei, Taiwan","institution_ids":["https://openalex.org/I134161618"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5021001528","display_name":"Po\u2010Yung Chou","orcid":null},"institutions":[{"id":"https://openalex.org/I134161618","display_name":"National Taiwan Normal University","ror":"https://ror.org/059dkdx38","country_code":"TW","type":"education","lineage":["https://openalex.org/I134161618"]}],"countries":["TW"],"is_corresponding":false,"raw_author_name":"Po-Yung Chou","raw_affiliation_strings":["Department of Electrical Engineering, National Taiwan Normal University, Taipei, Taiwan"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Department of Electrical Engineering, National Taiwan Normal University, Taipei, Taiwan","institution_ids":["https://openalex.org/I134161618"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5135895957","display_name":"Cheng-Hung Lin","orcid":"https://orcid.org/0000-0003-0044-3840"},"institutions":[{"id":"https://openalex.org/I134161618","display_name":"National Taiwan Normal University","ror":"https://ror.org/059dkdx38","country_code":"TW","type":"education","lineage":["https://openalex.org/I134161618"]}],"countries":["TW"],"is_corresponding":false,"raw_author_name":"Cheng-Hung Lin","raw_affiliation_strings":["Department of Electrical Engineering, National Taiwan Normal University, Taipei, Taiwan"],"raw_orcid":"https://orcid.org/0000-0003-0044-3840","affiliations":[{"raw_affiliation_string":"Department of Electrical Engineering, National Taiwan Normal University, Taipei, Taiwan","institution_ids":["https://openalex.org/I134161618"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":3,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":{"value":1850,"currency":"USD","value_usd":1850},"apc_paid":{"value":1850,"currency":"USD","value_usd":1850},"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.73675238,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":"14","issue":null,"first_page":"74912","last_page":"74922"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T12072","display_name":"Machine Learning and Algorithms","score":0.08320000022649765,"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/T12072","display_name":"Machine Learning and Algorithms","score":0.08320000022649765,"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/T12761","display_name":"Data Stream Mining Techniques","score":0.06440000236034393,"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/T12205","display_name":"Time Series Analysis and Forecasting","score":0.04410000145435333,"subfield":{"id":"https://openalex.org/subfields/1711","display_name":"Signal Processing"},"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/stability","display_name":"Stability (learning theory)","score":0.6644999980926514},{"id":"https://openalex.org/keywords/pruning","display_name":"Pruning","score":0.6025000214576721},{"id":"https://openalex.org/keywords/sampling","display_name":"Sampling (signal processing)","score":0.5338000059127808},{"id":"https://openalex.org/keywords/data-modeling","display_name":"Data modeling","score":0.38429999351501465},{"id":"https://openalex.org/keywords/pattern-recognition","display_name":"Pattern recognition (psychology)","score":0.3416000008583069},{"id":"https://openalex.org/keywords/sampled-data-systems","display_name":"Sampled data systems","score":0.3188999891281128}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.671500027179718},{"id":"https://openalex.org/C112972136","wikidata":"https://www.wikidata.org/wiki/Q7595718","display_name":"Stability (learning theory)","level":2,"score":0.6644999980926514},{"id":"https://openalex.org/C108010975","wikidata":"https://www.wikidata.org/wiki/Q500094","display_name":"Pruning","level":2,"score":0.6025000214576721},{"id":"https://openalex.org/C140779682","wikidata":"https://www.wikidata.org/wiki/Q210868","display_name":"Sampling (signal processing)","level":3,"score":0.5338000059127808},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.44519999623298645},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.43209999799728394},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.40950000286102295},{"id":"https://openalex.org/C67186912","wikidata":"https://www.wikidata.org/wiki/Q367664","display_name":"Data modeling","level":2,"score":0.38429999351501465},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.3416000008583069},{"id":"https://openalex.org/C90794131","wikidata":"https://www.wikidata.org/wiki/Q7410146","display_name":"Sampled data systems","level":3,"score":0.3188999891281128},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.28870001435279846},{"id":"https://openalex.org/C106516650","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm design","level":2,"score":0.287200003862381},{"id":"https://openalex.org/C2985946229","wikidata":"https://www.wikidata.org/wiki/Q49908","display_name":"Data sampling","level":2,"score":0.28700000047683716},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.2799000144004822},{"id":"https://openalex.org/C52740198","wikidata":"https://www.wikidata.org/wiki/Q1539564","display_name":"Importance sampling","level":3,"score":0.27549999952316284},{"id":"https://openalex.org/C2778755073","wikidata":"https://www.wikidata.org/wiki/Q10858537","display_name":"Scale (ratio)","level":2,"score":0.2648000121116638},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.2632000148296356},{"id":"https://openalex.org/C179799912","wikidata":"https://www.wikidata.org/wiki/Q205084","display_name":"Computational complexity theory","level":2,"score":0.25600001215934753},{"id":"https://openalex.org/C3018263672","wikidata":"https://www.wikidata.org/wiki/Q1296251","display_name":"Efficient algorithm","level":2,"score":0.2538999915122986}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1109/access.2026.3692050","is_oa":true,"landing_page_url":"https://doi.org/10.1109/access.2026.3692050","pdf_url":null,"source":{"id":"https://openalex.org/S2485537415","display_name":"IEEE Access","issn_l":"2169-3536","issn":["2169-3536"],"is_oa":true,"is_in_doaj":true,"is_core":true,"host_organization":"https://openalex.org/P4310319808","host_organization_name":"Institute of Electrical and Electronics Engineers","host_organization_lineage":["https://openalex.org/P4310319808"],"host_organization_lineage_names":["Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Access","raw_type":"journal-article"},{"id":"pmh:oai:doaj.org/article:ad1c4ab84e59433d82ba0f27fd3baf3f","is_oa":true,"landing_page_url":"https://doaj.org/article/ad1c4ab84e59433d82ba0f27fd3baf3f","pdf_url":null,"source":{"id":"https://openalex.org/S4306401280","display_name":"DOAJ (DOAJ: Directory of Open Access Journals)","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":"cc-by-sa","license_id":"https://openalex.org/licenses/cc-by-sa","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"IEEE Access, Vol 14, Pp 74912-74922 (2026)","raw_type":"article"}],"best_oa_location":{"id":"doi:10.1109/access.2026.3692050","is_oa":true,"landing_page_url":"https://doi.org/10.1109/access.2026.3692050","pdf_url":null,"source":{"id":"https://openalex.org/S2485537415","display_name":"IEEE Access","issn_l":"2169-3536","issn":["2169-3536"],"is_oa":true,"is_in_doaj":true,"is_core":true,"host_organization":"https://openalex.org/P4310319808","host_organization_name":"Institute of Electrical and Electronics Engineers","host_organization_lineage":["https://openalex.org/P4310319808"],"host_organization_lineage_names":["Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Access","raw_type":"journal-article"},"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":{"Recently,":[0],"deep":[1],"learning":[2,44],"has":[3,53],"achieved":[4],"remarkable":[5],"performance":[6,138],"gains":[7],"across":[8,190],"a":[9,132,147,164],"wide":[10],"range":[11],"of":[12,68,74,94,126],"vision":[13],"tasks,":[14],"largely":[15],"driven":[16],"by":[17,64,159],"increasingly":[18],"large-scale":[19],"training":[20,62,69,97,211],"datasets.":[21],"However,":[22],"recent":[23],"studies":[24],"have":[25],"shown":[26],"that":[27,180],"many":[28],"datasets":[29],"contain":[30],"redundant":[31],"or":[32],"even":[33],"mislabeled":[34],"samples,":[35],"which":[36,89,114],"not":[37],"only":[38],"provide":[39],"limited":[40],"benefit":[41],"for":[42,60],"model":[43],"but":[45],"also":[46],"incur":[47],"substantial":[48],"computational":[49,82],"costs.":[50],"Data":[51],"pruning":[52,77,108,142,149,188,192],"therefore":[54],"emerged":[55],"as":[56],"an":[57,105],"effective":[58],"strategy":[59],"improving":[61],"efficiency":[63],"selecting":[65],"representative":[66],"subsets":[67],"data.":[70],"A":[71],"major":[72],"limitation":[73],"existing":[75,186],"data":[76,107],"methods":[78],"is":[79,220],"the":[80,91,123,151,155,169,181,201],"high":[81,141],"overhead":[83],"required":[84],"to":[85,136],"compute":[86],"sample":[87,116,156],"scores,":[88],"contradicts":[90],"original":[92],"goal":[93],"reducing":[95],"overall":[96,210],"cost.":[98],"To":[99],"address":[100],"this":[101],"issue,":[102],"we":[103,130],"propose":[104],"efficient":[106],"framework":[109],"called":[110],"Early-Stage":[111],"Prediction":[112],"Stability,":[113],"evaluates":[115],"scores":[117],"based":[118],"on":[119,175,199],"prediction":[120],"stability":[121],"during":[122],"early":[124],"stages":[125],"training.":[127],"In":[128],"addition,":[129],"introduce":[131],"Balanced":[133],"Sampling":[134],"Strategy":[135],"prevent":[137],"degradation":[139],"under":[140,146,195],"ratios.":[143,193],"On":[144],"CIFAR-10,":[145],"90%":[148],"ratio,":[150],"proposed":[152,182,202],"method":[153,183,203],"reduces":[154,209],"scoring":[157],"time":[158,212],"approximately":[160],"66%":[161],"while":[162,213],"achieving":[163,214],"+4.19%":[165],"accuracy":[166],"improvement":[167],"over":[168],"previous":[170],"state-of-the-art":[171,187,215],"method.":[172],"Extensive":[173],"experiments":[174],"CIFAR-10":[176],"and":[177,208],"CIFAR-100":[178],"demonstrate":[179],"consistently":[184],"outperforms":[185],"approaches":[189],"multiple":[191],"Furthermore,":[194],"20%":[196],"label":[197],"noise":[198],"Tiny-ImageNet,":[200],"effectively":[204],"removes":[205],"noisy":[206],"samples":[207],"accuracy.":[216],"The":[217],"source":[218],"code":[219],"publicly":[221],"available":[222],"at":[223],"https://github.com/Hu210/esps-data-pruning.":[224]},"counts_by_year":[],"updated_date":"2026-06-11T09:08:48.828518","created_date":"2026-05-12T00:00:00"}