{"id":"https://openalex.org/W4416956239","doi":"https://doi.org/10.1145/3769102.3774378","title":"Poster: A Lightweight Pruning for Mitigating Neural Network Backdoor on Edge","display_name":"Poster: A Lightweight Pruning for Mitigating Neural Network Backdoor on Edge","publication_year":2025,"publication_date":"2025-12-03","ids":{"openalex":"https://openalex.org/W4416956239","doi":"https://doi.org/10.1145/3769102.3774378"},"language":null,"primary_location":{"id":"doi:10.1145/3769102.3774378","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3769102.3774378","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3769102.3774378","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the Tenth ACM/IEEE Symposium on Edge Computing","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://dl.acm.org/doi/pdf/10.1145/3769102.3774378","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5120581939","display_name":"Gabriel Wolff Taveira","orcid":"https://orcid.org/0009-0002-5696-7985"},"institutions":[{"id":"https://openalex.org/I43579087","display_name":"University of Wisconsin\u2013Milwaukee","ror":"https://ror.org/031q21x57","country_code":"US","type":"education","lineage":["https://openalex.org/I43579087"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Gabriel Wolff Taveira","raw_affiliation_strings":["University of Wisconsin-Milwaukee, Milwaukee, USA"],"raw_orcid":"https://orcid.org/0009-0002-5696-7985","affiliations":[{"raw_affiliation_string":"University of Wisconsin-Milwaukee, Milwaukee, USA","institution_ids":["https://openalex.org/I43579087"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Zijian Zhang","orcid":"https://orcid.org/0009-0009-9122-1637"},"institutions":[{"id":"https://openalex.org/I43579087","display_name":"University of Wisconsin\u2013Milwaukee","ror":"https://ror.org/031q21x57","country_code":"US","type":"education","lineage":["https://openalex.org/I43579087"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Zijian Zhang","raw_affiliation_strings":["University of Wisconsin-Milwaukee, Milwaukee, USA"],"raw_orcid":"https://orcid.org/0009-0009-9122-1637","affiliations":[{"raw_affiliation_string":"University of Wisconsin-Milwaukee, Milwaukee, USA","institution_ids":["https://openalex.org/I43579087"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100360185","display_name":"Zhen Zeng","orcid":"https://orcid.org/0000-0002-7877-292X"},"institutions":[{"id":"https://openalex.org/I43579087","display_name":"University of Wisconsin\u2013Milwaukee","ror":"https://ror.org/031q21x57","country_code":"US","type":"education","lineage":["https://openalex.org/I43579087"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Zhen Zeng","raw_affiliation_strings":["University of Wisconsin-Milwaukee, Milwaukee, USA"],"raw_orcid":"https://orcid.org/0000-0002-7877-292X","affiliations":[{"raw_affiliation_string":"University of Wisconsin-Milwaukee, Milwaukee, USA","institution_ids":["https://openalex.org/I43579087"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5057478629","display_name":"Zhongshu Gu","orcid":"https://orcid.org/0000-0001-9624-2669"},"institutions":[{"id":"https://openalex.org/I1341412227","display_name":"IBM (United States)","ror":"https://ror.org/05hh8d621","country_code":"US","type":"company","lineage":["https://openalex.org/I1341412227"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Zhongshu Gu","raw_affiliation_strings":["IBM Research, Yorktown Heights, USA"],"raw_orcid":"https://orcid.org/0000-0001-9624-2669","affiliations":[{"raw_affiliation_string":"IBM Research, Yorktown Heights, USA","institution_ids":["https://openalex.org/I1341412227"]}]}],"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":0,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"3"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.9854000210762024,"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/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.9854000210762024,"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":0.002899999963119626,"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/T12026","display_name":"Explainable Artificial Intelligence (XAI)","score":0.0013000000035390258,"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/backdoor","display_name":"Backdoor","score":0.9778000116348267},{"id":"https://openalex.org/keywords/pruning","display_name":"Pruning","score":0.7986999750137329},{"id":"https://openalex.org/keywords/exploit","display_name":"Exploit","score":0.6610999703407288},{"id":"https://openalex.org/keywords/enhanced-data-rates-for-gsm-evolution","display_name":"Enhanced Data Rates for GSM Evolution","score":0.5842999815940857},{"id":"https://openalex.org/keywords/reduction","display_name":"Reduction (mathematics)","score":0.5181000232696533},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.5126000046730042},{"id":"https://openalex.org/keywords/overhead","display_name":"Overhead (engineering)","score":0.44519999623298645}],"concepts":[{"id":"https://openalex.org/C2781045450","wikidata":"https://www.wikidata.org/wiki/Q254569","display_name":"Backdoor","level":2,"score":0.9778000116348267},{"id":"https://openalex.org/C108010975","wikidata":"https://www.wikidata.org/wiki/Q500094","display_name":"Pruning","level":2,"score":0.7986999750137329},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7215999960899353},{"id":"https://openalex.org/C165696696","wikidata":"https://www.wikidata.org/wiki/Q11287","display_name":"Exploit","level":2,"score":0.6610999703407288},{"id":"https://openalex.org/C162307627","wikidata":"https://www.wikidata.org/wiki/Q204833","display_name":"Enhanced Data Rates for GSM Evolution","level":2,"score":0.5842999815940857},{"id":"https://openalex.org/C111335779","wikidata":"https://www.wikidata.org/wiki/Q3454686","display_name":"Reduction (mathematics)","level":2,"score":0.5181000232696533},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.5126000046730042},{"id":"https://openalex.org/C2779960059","wikidata":"https://www.wikidata.org/wiki/Q7113681","display_name":"Overhead (engineering)","level":2,"score":0.44519999623298645},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3885999917984009},{"id":"https://openalex.org/C126780896","wikidata":"https://www.wikidata.org/wiki/Q899871","display_name":"Distortion (music)","level":4,"score":0.350600004196167},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.34060001373291016},{"id":"https://openalex.org/C2984842247","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep neural networks","level":3,"score":0.3292999863624573},{"id":"https://openalex.org/C138236772","wikidata":"https://www.wikidata.org/wiki/Q25098575","display_name":"Edge device","level":3,"score":0.3222000002861023},{"id":"https://openalex.org/C3770464","wikidata":"https://www.wikidata.org/wiki/Q775963","display_name":"Smoothing","level":2,"score":0.29409998655319214},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.29350000619888306},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.2685999870300293},{"id":"https://openalex.org/C28719098","wikidata":"https://www.wikidata.org/wiki/Q44946","display_name":"Point (geometry)","level":2,"score":0.25}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3769102.3774378","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3769102.3774378","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3769102.3774378","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the Tenth ACM/IEEE Symposium on Edge Computing","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.1145/3769102.3774378","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3769102.3774378","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3769102.3774378","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the Tenth ACM/IEEE Symposium on Edge Computing","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":false},"content_urls":{"pdf":"https://content.openalex.org/works/W4416956239.pdf"},"referenced_works_count":10,"referenced_works":["https://openalex.org/W2145287260","https://openalex.org/W2194775991","https://openalex.org/W2807363941","https://openalex.org/W2942091739","https://openalex.org/W2963363373","https://openalex.org/W2996800219","https://openalex.org/W3088144810","https://openalex.org/W4386147982","https://openalex.org/W4390871934","https://openalex.org/W6926330407"],"related_works":[],"abstract_inverted_index":{"Image-based":[0],"neural":[1],"networks":[2],"are":[3,103],"widely":[4],"used":[5],"across":[6],"diverse":[7],"applications,":[8],"yet":[9],"their":[10],"inherent":[11],"susceptibility":[12],"to":[13,31,65,97,127,132,167,185],"backdoor":[14,23,140],"attacks":[15],"has":[16],"raised":[17],"growing":[18],"concerns.":[19],"In":[20,71],"particular,":[21],"data-free":[22],"attacks,":[24],"which":[25,54],"exploit":[26],"benign":[27],"but":[28,130],"redundant":[29],"pathways":[30],"manipulate":[32],"model":[33,117,146],"predictions,":[34],"pose":[35],"significant":[36],"risks.":[37],"Existing":[38],"defense":[39,78],"methods":[40],"rely":[41],"heavily":[42],"on":[43,48],"conventional":[44,186],"pruning":[45,188,191],"strategies":[46],"based":[47],"weight":[49],"magnitude":[50],"or":[51,115],"neuron":[52],"activation,":[53],"often":[55],"cause":[56],"excessive":[57],"distortion":[58],"of":[59,68,178,181],"the":[60,182],"original":[61],"representations":[62],"and":[63,105,175,189],"lead":[64],"substantial":[66],"loss":[67],"useful":[69],"information.":[70],"this":[72],"work,":[73],"we":[74,120],"propose":[75],"a":[76,83,136,149,159,171],"pruning-based":[77],"method":[79,87],"ALobot":[80,157],"that":[81,124,156],"leverages":[82],"label-dependence":[84],"pruning.":[85],"This":[86],"is":[88],"especially":[89],"well-suited":[90],"for":[91],"mitigating":[92],"backdoors":[93],"in":[94,139,161],"models":[95],"deployed":[96],"edge":[98],"devices,":[99],"where":[100],"computational":[101],"resources":[102],"constrained":[104],"defenses":[106],"must":[107],"lower":[108],"attack":[109,141,162],"success":[110,142,163],"without":[111,144],"imposing":[112],"heavy":[113],"overhead":[114],"degrading":[116],"performance.":[118],"Specifically,":[119],"selectively":[121],"prune":[122],"neurons":[123],"contribute":[125],"substantially":[126],"inter-class":[128],"discrimination":[129],"minimally":[131],"intra-class":[133],"consistency,":[134],"achieving":[135],"noticeable":[137],"reduction":[138,160],"rate":[143,164],"sacrificing":[145],"performance":[147,174],"through":[148],"lightweight":[150],"mitigation.":[151],"The":[152],"initial":[153],"results":[154],"show":[155],"achieves":[158],"by":[165],"2%":[166],"8%,":[168],"while":[169],"maintaining":[170],"high":[172],"clean":[173],"requiring":[176],"modification":[177],"only":[179],"1/20":[180],"parameters":[183],"compared":[184],"L1-norm":[187],"L2-norm":[190],"methods.":[192]},"counts_by_year":[],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-12-03T00:00:00"}