{"id":"https://openalex.org/W6966821887","doi":"https://doi.org/10.48550/arxiv.2506.17621","title":"Exploiting Efficiency Vulnerabilities in Dynamic Deep Learning Systems","display_name":"Exploiting Efficiency Vulnerabilities in Dynamic Deep Learning Systems","publication_year":2025,"publication_date":"2025-06-21","ids":{"openalex":"https://openalex.org/W6966821887","doi":"https://doi.org/10.48550/arxiv.2506.17621"},"language":"en","primary_location":{"id":"doi:10.48550/arxiv.2506.17621","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2506.17621","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":"Preprint"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2506.17621","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":null,"display_name":"Rathnasuriya, Ravishka","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Rathnasuriya, Ravishka","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":null,"display_name":"Yang, Wei","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yang, Wei","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"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":true,"primary_topic":{"id":"https://openalex.org/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.9901000261306763,"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":0.9901000261306763,"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/T11424","display_name":"Security and Verification in Computing","score":0.0017000000225380063,"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":0.0013000000035390258,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/nucleofection","display_name":"Nucleofection","score":0.39079999923706055},{"id":"https://openalex.org/keywords/gestational-period","display_name":"Gestational period","score":0.3621000051498413},{"id":"https://openalex.org/keywords/proteogenomics","display_name":"Proteogenomics","score":0.34450000524520874},{"id":"https://openalex.org/keywords/tsg101","display_name":"TSG101","score":0.33180001378059387},{"id":"https://openalex.org/keywords/hyporeflexia","display_name":"Hyporeflexia","score":0.32120001316070557},{"id":"https://openalex.org/keywords/articular-cartilage-damage","display_name":"Articular cartilage damage","score":0.31130000948905945},{"id":"https://openalex.org/keywords/hemopericardium","display_name":"Hemopericardium","score":0.2903999984264374},{"id":"https://openalex.org/keywords/diafiltration","display_name":"Diafiltration","score":0.28850001096725464}],"concepts":[{"id":"https://openalex.org/C144251240","wikidata":"https://www.wikidata.org/wiki/Q7068229","display_name":"Nucleofection","level":4,"score":0.39079999923706055},{"id":"https://openalex.org/C2992336715","wikidata":"https://www.wikidata.org/wiki/Q63431143","display_name":"Gestational period","level":4,"score":0.3621000051498413},{"id":"https://openalex.org/C145741570","wikidata":"https://www.wikidata.org/wiki/Q7251534","display_name":"Proteogenomics","level":5,"score":0.34450000524520874},{"id":"https://openalex.org/C2778283623","wikidata":"https://www.wikidata.org/wiki/Q18032200","display_name":"TSG101","level":5,"score":0.33180001378059387},{"id":"https://openalex.org/C2777158700","wikidata":"https://www.wikidata.org/wiki/Q1419356","display_name":"Hyporeflexia","level":3,"score":0.32120001316070557},{"id":"https://openalex.org/C2781032047","wikidata":"https://www.wikidata.org/wiki/Q938793","display_name":"Articular cartilage damage","level":5,"score":0.31130000948905945},{"id":"https://openalex.org/C2777935831","wikidata":"https://www.wikidata.org/wiki/Q3144949","display_name":"Hemopericardium","level":3,"score":0.2903999984264374},{"id":"https://openalex.org/C18743360","wikidata":"https://www.wikidata.org/wiki/Q1208096","display_name":"Diafiltration","level":4,"score":0.28850001096725464},{"id":"https://openalex.org/C2779627259","wikidata":"https://www.wikidata.org/wiki/Q779763","display_name":"Pretext","level":3,"score":0.288100004196167},{"id":"https://openalex.org/C133074676","wikidata":"https://www.wikidata.org/wiki/Q428729","display_name":"Fusible alloy","level":2,"score":0.2842000126838684},{"id":"https://openalex.org/C2777054765","wikidata":"https://www.wikidata.org/wiki/Q6402731","display_name":"Dysgeusia","level":3,"score":0.27790001034736633},{"id":"https://openalex.org/C180938184","wikidata":"https://www.wikidata.org/wiki/Q2142270","display_name":"Liquation","level":3,"score":0.2766000032424927},{"id":"https://openalex.org/C135979968","wikidata":"https://www.wikidata.org/wiki/Q609809","display_name":"Protein isoform","level":5,"score":0.27570000290870667},{"id":"https://openalex.org/C2776356786","wikidata":"https://www.wikidata.org/wiki/Q1048573","display_name":"Tubulopathy","level":3,"score":0.26919999718666077},{"id":"https://openalex.org/C2777742743","wikidata":"https://www.wikidata.org/wiki/Q19904005","display_name":"Durvalumab","level":5,"score":0.2614000141620636},{"id":"https://openalex.org/C2909186138","wikidata":"https://www.wikidata.org/wiki/Q1500373","display_name":"Hyperlactatemia","level":2,"score":0.2599000036716461},{"id":"https://openalex.org/C104545631","wikidata":"https://www.wikidata.org/wiki/Q464858","display_name":"Demotion","level":3,"score":0.2526000142097473}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2506.17621","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2506.17621","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":"Preprint"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2506.17621","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2506.17621","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":"Preprint"},"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/7","display_name":"Affordable and clean energy","score":0.5654830932617188}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"The":[0],"growing":[1],"deployment":[2],"of":[3,87,107,117,137],"deep":[4,28],"learning":[5,29],"models":[6],"in":[7,45,70,78,90,114,123],"real-world":[8],"environments":[9],"has":[10],"intensified":[11],"the":[12,84,115,135],"need":[13],"for":[14,64],"efficient":[15],"inference":[16],"under":[17,150],"strict":[18],"latency":[19],"and":[20,53,75,92,121,143],"resource":[21],"constraints.":[22],"To":[23],"meet":[24],"these":[25,42,129],"demands,":[26],"dynamic":[27,49,88],"systems":[30,43,96],"(DDLSs)":[31],"have":[32],"emerged,":[33],"offering":[34],"input-adaptive":[35],"computation":[36],"to":[37,66,133,147],"optimize":[38],"runtime":[39],"efficiency.":[40],"While":[41],"succeed":[44],"reducing":[46],"cost,":[47],"their":[48],"nature":[50],"introduces":[51],"subtle":[52],"underexplored":[54],"security":[55,85],"risks.":[56],"In":[57],"particular,":[58],"input-dependent":[59],"execution":[60],"pathways":[61],"create":[62],"opportunities":[63],"adversaries":[65],"degrade":[67],"efficiency,":[68],"resulting":[69],"excessive":[71],"latency,":[72],"energy":[73],"usage,":[74],"potential":[76],"denial-of-service":[77],"time-sensitive":[79],"deployments.":[80],"This":[81],"work":[82],"investigates":[83],"implications":[86],"behaviors":[89],"DDLSs":[91,142],"reveals":[93],"how":[94],"current":[95,124],"expose":[97],"efficiency":[98,138],"vulnerabilities":[99],"exploitable":[100],"by":[101],"adversarial":[102,151],"inputs.":[103],"Through":[104],"a":[105],"survey":[106],"existing":[108],"attack":[109],"strategies,":[110],"we":[111,131],"identify":[112],"gaps":[113],"coverage":[116],"emerging":[118],"model":[119],"architectures":[120],"limitations":[122],"defense":[125],"mechanisms.":[126],"Building":[127],"on":[128,140],"insights,":[130],"propose":[132],"examine":[134],"feasibility":[136],"attacks":[139],"modern":[141],"develop":[144],"targeted":[145],"defenses":[146],"preserve":[148],"robustness":[149],"conditions.":[152]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2025-10-10T00:00:00"}