{"id":"https://openalex.org/W7154305441","doi":"https://doi.org/10.48550/arxiv.2604.09994","title":"Aging Aware Adaptive Voltage Scaling for Reliable and Efficient AI Accelerators","display_name":"Aging Aware Adaptive Voltage Scaling for Reliable and Efficient AI Accelerators","publication_year":2026,"publication_date":"2026-04-11","ids":{"openalex":"https://openalex.org/W7154305441","doi":"https://doi.org/10.48550/arxiv.2604.09994"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2604.09994","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.09994","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.2604.09994","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5133626643","display_name":"Tong Xie","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xie, Tong","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5133607813","display_name":"Zuodong Zhang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhang, Zuodong","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5133568525","display_name":"Chao Yang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yang, Chao","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5133610910","display_name":"Yuan Wang","orcid":"https://orcid.org/0009-0009-8136-4255"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Yuan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5133624681","display_name":"Runsheng Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Runsheng","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5133570818","display_name":"Meng Li","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Li, Meng","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":false,"primary_topic":{"id":"https://openalex.org/T10502","display_name":"Advanced Memory and Neural Computing","score":0.24379999935626984,"subfield":{"id":"https://openalex.org/subfields/2208","display_name":"Electrical and Electronic Engineering"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T10502","display_name":"Advanced Memory and Neural Computing","score":0.24379999935626984,"subfield":{"id":"https://openalex.org/subfields/2208","display_name":"Electrical and Electronic Engineering"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.23909999430179596,"subfield":{"id":"https://openalex.org/subfields/1708","display_name":"Hardware and Architecture"},"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/T10472","display_name":"Semiconductor materials and devices","score":0.12680000066757202,"subfield":{"id":"https://openalex.org/subfields/2208","display_name":"Electrical and Electronic Engineering"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/extrapolation","display_name":"Extrapolation","score":0.579800009727478},{"id":"https://openalex.org/keywords/resilience","display_name":"Resilience (materials science)","score":0.5364999771118164},{"id":"https://openalex.org/keywords/voltage","display_name":"Voltage","score":0.5217999815940857},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.5120000243186951},{"id":"https://openalex.org/keywords/exploit","display_name":"Exploit","score":0.4805999994277954},{"id":"https://openalex.org/keywords/reboot","display_name":"Reboot","score":0.4499000012874603},{"id":"https://openalex.org/keywords/dram","display_name":"Dram","score":0.42649999260902405},{"id":"https://openalex.org/keywords/benchmark","display_name":"Benchmark (surveying)","score":0.41269999742507935},{"id":"https://openalex.org/keywords/power","display_name":"Power (physics)","score":0.40380001068115234}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6438000202178955},{"id":"https://openalex.org/C132459708","wikidata":"https://www.wikidata.org/wiki/Q744069","display_name":"Extrapolation","level":2,"score":0.579800009727478},{"id":"https://openalex.org/C2779585090","wikidata":"https://www.wikidata.org/wiki/Q3457762","display_name":"Resilience (materials science)","level":2,"score":0.5364999771118164},{"id":"https://openalex.org/C165801399","wikidata":"https://www.wikidata.org/wiki/Q25428","display_name":"Voltage","level":2,"score":0.5217999815940857},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.5120000243186951},{"id":"https://openalex.org/C165696696","wikidata":"https://www.wikidata.org/wiki/Q11287","display_name":"Exploit","level":2,"score":0.4805999994277954},{"id":"https://openalex.org/C120524526","wikidata":"https://www.wikidata.org/wiki/Q1709148","display_name":"Reboot","level":2,"score":0.4499000012874603},{"id":"https://openalex.org/C7366592","wikidata":"https://www.wikidata.org/wiki/Q1255620","display_name":"Dram","level":2,"score":0.42649999260902405},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.41269999742507935},{"id":"https://openalex.org/C163258240","wikidata":"https://www.wikidata.org/wiki/Q25342","display_name":"Power (physics)","level":2,"score":0.40380001068115234},{"id":"https://openalex.org/C99844830","wikidata":"https://www.wikidata.org/wiki/Q102441924","display_name":"Scaling","level":2,"score":0.40220001339912415},{"id":"https://openalex.org/C200601418","wikidata":"https://www.wikidata.org/wiki/Q2193887","display_name":"Reliability engineering","level":1,"score":0.38119998574256897},{"id":"https://openalex.org/C157742956","wikidata":"https://www.wikidata.org/wiki/Q3237776","display_name":"Frequency scaling","level":3,"score":0.33500000834465027},{"id":"https://openalex.org/C63479239","wikidata":"https://www.wikidata.org/wiki/Q7353546","display_name":"Robustness (evolution)","level":3,"score":0.3249000012874603},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.3082999885082245},{"id":"https://openalex.org/C40760162","wikidata":"https://www.wikidata.org/wiki/Q10920295","display_name":"Voltage droop","level":4,"score":0.30640000104904175},{"id":"https://openalex.org/C168292644","wikidata":"https://www.wikidata.org/wiki/Q10860336","display_name":"Power optimization","level":4,"score":0.2849000096321106},{"id":"https://openalex.org/C2777062904","wikidata":"https://www.wikidata.org/wiki/Q545406","display_name":"Toolchain","level":3,"score":0.2847999930381775},{"id":"https://openalex.org/C2742236","wikidata":"https://www.wikidata.org/wiki/Q924713","display_name":"Efficient energy use","level":2,"score":0.28119999170303345},{"id":"https://openalex.org/C24326235","wikidata":"https://www.wikidata.org/wiki/Q126095","display_name":"Electronic engineering","level":1,"score":0.27559998631477356},{"id":"https://openalex.org/C79403827","wikidata":"https://www.wikidata.org/wiki/Q3988","display_name":"Real-time computing","level":1,"score":0.2752000093460083},{"id":"https://openalex.org/C41661131","wikidata":"https://www.wikidata.org/wiki/Q220764","display_name":"Interrupt","level":3,"score":0.2745000123977661},{"id":"https://openalex.org/C2984842247","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep neural networks","level":3,"score":0.27239999175071716},{"id":"https://openalex.org/C2779679103","wikidata":"https://www.wikidata.org/wiki/Q5251805","display_name":"Degradation (telecommunications)","level":2,"score":0.2694999873638153},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.26840001344680786},{"id":"https://openalex.org/C557185","wikidata":"https://www.wikidata.org/wiki/Q6987194","display_name":"Negative-bias temperature instability","level":5,"score":0.26809999346733093},{"id":"https://openalex.org/C157764524","wikidata":"https://www.wikidata.org/wiki/Q1383412","display_name":"Throughput","level":3,"score":0.26159998774528503},{"id":"https://openalex.org/C128624480","wikidata":"https://www.wikidata.org/wiki/Q1504817","display_name":"Low voltage","level":3,"score":0.2572000026702881},{"id":"https://openalex.org/C88182573","wikidata":"https://www.wikidata.org/wiki/Q1139740","display_name":"High voltage","level":3,"score":0.25110000371932983}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2604.09994","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.09994","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.2604.09994","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.09994","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":[{"score":0.8249387741088867,"display_name":"Affordable and clean energy","id":"https://metadata.un.org/sdg/7"}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Deep":[0],"neural":[1],"networks":[2],"(DNNs)":[3],"have":[4],"showcased":[5],"remarkable":[6],"performance":[7],"across":[8],"various":[9],"tasks":[10],"and":[11,32,72,82,89,109,130,156,179],"are":[12,35],"widely":[13],"deployed":[14],"on":[15,48,116,163],"AI":[16,91],"accelerators":[17],"fabricated":[18],"in":[19,95],"advanced":[20],"technology":[21],"nodes":[22],"for":[23,112,154,158],"efficiency.":[24],"As":[25],"aging":[26,50,81,102,172],"effects":[27,108],"become":[28],"more":[29],"pronounced,":[30],"timing":[31],"voltage":[33,40,46,77,123,132,149],"guardbands":[34],"increasingly":[36],"applied.":[37],"Aging-aware":[38],"adaptive":[39],"scaling":[41,124],"(AVS),":[42],"which":[43],"adjusts":[44],"supply":[45,76],"based":[47],"on-chip":[49],"scenarios,":[51],"has":[52],"emerged":[53],"as":[54],"a":[55,121],"promising":[56],"solution":[57],"to":[58,176,190],"avoid":[59],"excessive":[60],"guardbanding.":[61],"However,":[62],"conventional":[63],"AVS":[64],"techniques":[65],"overlook":[66],"the":[67,75,141],"inherent":[68],"resilience":[69,129],"of":[70,143],"DNNs":[71],"frequently":[73],"raise":[74],"unnecessarily,":[78],"thereby":[79],"exacerbating":[80],"increasing":[83],"power":[84,187],"consumption.":[85],"To":[86],"enable":[87],"reliable":[88],"efficient":[90],"inference":[92],"with":[93],"AVS,":[94],"this":[96,117],"paper,":[97],"we":[98,119],"develop":[99],"an":[100],"accurate":[101],"prediction":[103],"framework":[104,139],"that":[105,126,137,168],"incorporates":[106],"historical":[107],"iterative":[110],"extrapolation":[111],"full-lifetime":[113],"modeling.":[114],"Building":[115],"framework,":[118],"propose":[120],"fault-tolerant":[122],"policy":[125],"exploits":[127],"DNN":[128,165],"defers":[131],"increases":[133],"accordingly.":[134],"Experiments":[135],"show":[136],"our":[138,169],"mitigates":[140],"pessimism":[142],"maximum-voltage":[144],"baselines,":[145],"reducing":[146],"predicted":[147],"threshold":[148],"shift":[150],"(\u0394Vth)":[151],"by":[152,174],"19.4%":[153],"PMOS":[155],"19.1%":[157],"NMOS,":[159],"respectively.":[160],"Furthermore,":[161],"evaluation":[162],"representative":[164],"workloads":[166],"demonstrates":[167],"optimization":[170],"reduces":[171],"degradation":[173],"up":[175],"45.8%":[177],"(NMOS)":[178],"30.6%":[180],"(PMOS)":[181],"while":[182],"achieving":[183],"14.0%":[184],"average":[185],"lifetime":[186],"savings":[188],"compared":[189],"resilience-agnostic":[191],"methods.":[192]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-04-15T00:00:00"}