{"id":"https://openalex.org/W7151492704","doi":"https://doi.org/10.48550/arxiv.2604.04681","title":"Batch Loss Score for Dynamic Data Pruning","display_name":"Batch Loss Score for Dynamic Data Pruning","publication_year":2026,"publication_date":"2026-04-06","ids":{"openalex":"https://openalex.org/W7151492704","doi":"https://doi.org/10.48550/arxiv.2604.04681"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2604.04681","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.04681","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.2604.04681","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5133127152","display_name":"Qing Zhou","orcid":null},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Zhou, Qing","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5113304218","display_name":"Bingxuan Zhao","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhao, Bingxuan","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5133064606","display_name":"Tao Yang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yang, Tao","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5133118905","display_name":"Hongyuan Zhang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhang, Hongyuan","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5133079954","display_name":"Junyu Gao","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Gao, Junyu","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5133070164","display_name":"Qi Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Qi","raw_affiliation_strings":[],"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":6,"corresponding_author_ids":["https://openalex.org/A5133127152"],"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/T10036","display_name":"Advanced Neural Network Applications","score":0.1842000037431717,"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"}},"topics":[{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.1842000037431717,"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/T12535","display_name":"Machine Learning and Data Classification","score":0.12860000133514404,"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/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.11140000075101852,"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/pruning","display_name":"Pruning","score":0.564300000667572},{"id":"https://openalex.org/keywords/code","display_name":"Code (set theory)","score":0.4986000061035156},{"id":"https://openalex.org/keywords/computation","display_name":"Computation","score":0.4796000123023987},{"id":"https://openalex.org/keywords/sample","display_name":"Sample (material)","score":0.47769999504089355},{"id":"https://openalex.org/keywords/noise","display_name":"Noise (video)","score":0.4327999949455261},{"id":"https://openalex.org/keywords/frame","display_name":"Frame (networking)","score":0.3774000108242035},{"id":"https://openalex.org/keywords/pattern-recognition","display_name":"Pattern recognition (psychology)","score":0.32199999690055847},{"id":"https://openalex.org/keywords/exponential-function","display_name":"Exponential function","score":0.30079999566078186}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.630299985408783},{"id":"https://openalex.org/C108010975","wikidata":"https://www.wikidata.org/wiki/Q500094","display_name":"Pruning","level":2,"score":0.564300000667572},{"id":"https://openalex.org/C2776760102","wikidata":"https://www.wikidata.org/wiki/Q5139990","display_name":"Code (set theory)","level":3,"score":0.4986000061035156},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.49079999327659607},{"id":"https://openalex.org/C45374587","wikidata":"https://www.wikidata.org/wiki/Q12525525","display_name":"Computation","level":2,"score":0.4796000123023987},{"id":"https://openalex.org/C198531522","wikidata":"https://www.wikidata.org/wiki/Q485146","display_name":"Sample (material)","level":2,"score":0.47769999504089355},{"id":"https://openalex.org/C99498987","wikidata":"https://www.wikidata.org/wiki/Q2210247","display_name":"Noise (video)","level":3,"score":0.4327999949455261},{"id":"https://openalex.org/C126042441","wikidata":"https://www.wikidata.org/wiki/Q1324888","display_name":"Frame (networking)","level":2,"score":0.3774000108242035},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3596999943256378},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.3467999994754791},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.337799996137619},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.32199999690055847},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.31850001215934753},{"id":"https://openalex.org/C151376022","wikidata":"https://www.wikidata.org/wiki/Q168698","display_name":"Exponential function","level":2,"score":0.30079999566078186},{"id":"https://openalex.org/C193519340","wikidata":"https://www.wikidata.org/wiki/Q891179","display_name":"Data loss","level":2,"score":0.29750001430511475},{"id":"https://openalex.org/C12713177","wikidata":"https://www.wikidata.org/wiki/Q1900281","display_name":"Perspective (graphical)","level":2,"score":0.2915000021457672},{"id":"https://openalex.org/C172658912","wikidata":"https://www.wikidata.org/wiki/Q661613","display_name":"Batch processing","level":2,"score":0.2872999906539917},{"id":"https://openalex.org/C2781170535","wikidata":"https://www.wikidata.org/wiki/Q30587856","display_name":"Noisy data","level":2,"score":0.2849999964237213},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.28349998593330383},{"id":"https://openalex.org/C2780148112","wikidata":"https://www.wikidata.org/wiki/Q1432581","display_name":"Proxy (statistics)","level":2,"score":0.2831999957561493},{"id":"https://openalex.org/C105795698","wikidata":"https://www.wikidata.org/wiki/Q12483","display_name":"Statistics","level":1,"score":0.28049999475479126},{"id":"https://openalex.org/C129848803","wikidata":"https://www.wikidata.org/wiki/Q2564360","display_name":"Sample size determination","level":2,"score":0.27790001034736633},{"id":"https://openalex.org/C160920958","wikidata":"https://www.wikidata.org/wiki/Q7662746","display_name":"Synthetic data","level":2,"score":0.26739999651908875},{"id":"https://openalex.org/C61797465","wikidata":"https://www.wikidata.org/wiki/Q1188986","display_name":"Term (time)","level":2,"score":0.26429998874664307},{"id":"https://openalex.org/C77618280","wikidata":"https://www.wikidata.org/wiki/Q1155772","display_name":"Scheme (mathematics)","level":2,"score":0.2603999972343445}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2604.04681","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.04681","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.2604.04681","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.04681","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":{"Dynamic":[0],"data":[1],"pruning":[2],"accelerates":[3],"deep":[4],"learning":[5],"by":[6,166],"selectively":[7],"omitting":[8],"less":[9],"informative":[10],"samples":[11,176],"during":[12],"training.":[13],"While":[14],"per-sample":[15,157,196],"loss":[16,33,197],"is":[17,98,164,198,203],"a":[18,48,78,82,107,118,134,140],"common":[19],"importance":[20],"metric,":[21],"obtaining":[22],"it":[23],"can":[24],"be":[25],"challenging":[26],"or":[27,32],"infeasible":[28],"for":[29,137,142,192],"complex":[30,193],"models":[31],"functions,":[34],"often":[35],"requiring":[36],"significant":[37],"implementation":[38],"effort.":[39],"This":[40,116],"work":[41],"proposes":[42],"the":[43,71,75,102,121,127,131],"Batch":[44],"Loss":[45],"Score":[46],"(BLS),":[47],"computationally":[49],"efficient":[50],"alternative":[51],"using":[52],"an":[53],"Exponential":[54],"Moving":[55],"Average":[56],"(EMA)":[57],"of":[58,77,85,126,175],"readily":[59,154],"available":[60,204],"batch":[61,72,95,113],"losses":[62],"to":[63,66,130,171,200],"assign":[64],"scores":[65],"individual":[67,88,128],"samples.":[68],"We":[69],"frame":[70],"loss,":[73,89,132],"from":[74,93],"perspective":[76],"single":[79],"sample,":[80],"as":[81,106,139],"noisy":[83],"measurement":[84],"its":[86,186],"scaled":[87],"with":[90],"noise":[91],"originating":[92],"stochastic":[94],"composition.":[96],"It":[97],"formally":[99],"shown":[100],"that":[101],"EMA":[103],"mechanism":[104],"functions":[105],"first-order":[108],"low-pass":[109],"filter,":[110],"attenuating":[111],"high-frequency":[112],"composition":[114],"noise.":[115],"yields":[117],"score":[119],"approximating":[120],"smoothed":[122],"and":[123,153,182,188],"persistent":[124],"contribution":[125],"sample":[129,143],"providing":[133],"theoretical":[135],"grounding":[136],"BLS":[138,145],"proxy":[141],"importance.":[144],"demonstrates":[146],"remarkable":[147],"code":[148],"integration":[149],"simplicity":[150],"(\\textbf{three-line":[151],"injection})":[152],"adapts":[155],"existing":[156],"loss-based":[158],"methods":[159,170],"(\\textbf{one-line":[160],"proxy}).":[161],"Its":[162],"effectiveness":[163],"demonstrated":[165],"enhancing":[167],"two":[168],"such":[169],"losslessly":[172],"prune":[173],"\\textbf{20\\%-50\\%}":[174],"across":[177],"\\textit{14":[178],"datasets},":[179],"\\textit{11":[180],"tasks}":[181],"\\textit{18":[183],"models},":[184],"highlighting":[185],"utility":[187],"broad":[189],"applicability,":[190],"especially":[191],"scenarios":[194],"where":[195],"difficult":[199],"access.":[201],"Code":[202],"at":[205],"https://github.com/mrazhou/BLS.":[206]},"counts_by_year":[],"updated_date":"2026-04-08T06:07:18.267832","created_date":"2026-04-08T00:00:00"}