{"id":"https://openalex.org/W7138045513","doi":"https://doi.org/10.1609/aaai.v40i33.40041","title":"PRISM: Privacy-Aware Routing for Adaptive Cloud\u2013Edge LLM Inference via Semantic Sketch Collaboration","display_name":"PRISM: Privacy-Aware Routing for Adaptive Cloud\u2013Edge LLM Inference via Semantic Sketch Collaboration","publication_year":2026,"publication_date":"2026-03-14","ids":{"openalex":"https://openalex.org/W7138045513","doi":"https://doi.org/10.1609/aaai.v40i33.40041"},"language":null,"primary_location":{"id":"doi:10.1609/aaai.v40i33.40041","is_oa":true,"landing_page_url":"https://doi.org/10.1609/aaai.v40i33.40041","pdf_url":null,"source":{"id":"https://openalex.org/S4210191458","display_name":"Proceedings of the AAAI Conference on Artificial Intelligence","issn_l":"2159-5399","issn":["2159-5399","2374-3468"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/P4310320058","host_organization_name":"Association for the Advancement of Artificial Intelligence","host_organization_lineage":["https://openalex.org/P4310320058"],"host_organization_lineage_names":["Association for the Advancement of Artificial Intelligence"],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the AAAI Conference on Artificial Intelligence","raw_type":"journal-article"},"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"diamond","oa_url":"https://doi.org/10.1609/aaai.v40i33.40041","any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5129750121","display_name":"Junfei Zhan","orcid":null},"institutions":[{"id":"https://openalex.org/I36788626","display_name":"California University of Pennsylvania","ror":"https://ror.org/01spssf70","country_code":"US","type":"education","lineage":["https://openalex.org/I36788626"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Junfei Zhan","raw_affiliation_strings":["University of Pennsylvania"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Pennsylvania","institution_ids":["https://openalex.org/I36788626"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5111301674","display_name":"Haoxun Shen","orcid":null},"institutions":[{"id":"https://openalex.org/I36788626","display_name":"California University of Pennsylvania","ror":"https://ror.org/01spssf70","country_code":"US","type":"education","lineage":["https://openalex.org/I36788626"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Haoxun Shen","raw_affiliation_strings":["University of Pennsylvania"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Pennsylvania","institution_ids":["https://openalex.org/I36788626"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5129659429","display_name":"Zheng Lin","orcid":null},"institutions":[{"id":"https://openalex.org/I889458895","display_name":"University of Hong Kong","ror":"https://ror.org/02zhqgq86","country_code":"HK","type":"education","lineage":["https://openalex.org/I889458895"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Zheng Lin","raw_affiliation_strings":["University of Hong Kong"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Hong Kong","institution_ids":["https://openalex.org/I889458895"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5129723807","display_name":"Tengjiao He","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Tengjiao He","raw_affiliation_strings":["Jinan University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Jinan University","institution_ids":[]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":1,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":"40","issue":"33","first_page":"28150","last_page":"28158"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10764","display_name":"Privacy-Preserving Technologies in Data","score":0.19850000739097595,"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/T10764","display_name":"Privacy-Preserving Technologies in Data","score":0.19850000739097595,"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/T10273","display_name":"IoT and Edge/Fog Computing","score":0.19439999759197235,"subfield":{"id":"https://openalex.org/subfields/1705","display_name":"Computer Networks and Communications"},"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/T14347","display_name":"Big Data and Digital Economy","score":0.12020000070333481,"subfield":{"id":"https://openalex.org/subfields/1710","display_name":"Information Systems"},"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/inference","display_name":"Inference","score":0.73089998960495},{"id":"https://openalex.org/keywords/differential-privacy","display_name":"Differential privacy","score":0.6751000285148621},{"id":"https://openalex.org/keywords/cloud-computing","display_name":"Cloud computing","score":0.629800021648407},{"id":"https://openalex.org/keywords/sketch","display_name":"Sketch","score":0.5989000201225281},{"id":"https://openalex.org/keywords/energy-consumption","display_name":"Energy consumption","score":0.5325999855995178},{"id":"https://openalex.org/keywords/overhead","display_name":"Overhead (engineering)","score":0.42590001225471497},{"id":"https://openalex.org/keywords/enhanced-data-rates-for-gsm-evolution","display_name":"Enhanced Data Rates for GSM Evolution","score":0.39980000257492065},{"id":"https://openalex.org/keywords/edge-device","display_name":"Edge device","score":0.3978999853134155},{"id":"https://openalex.org/keywords/latency","display_name":"Latency (audio)","score":0.36959999799728394}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8468999862670898},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.73089998960495},{"id":"https://openalex.org/C23130292","wikidata":"https://www.wikidata.org/wiki/Q5275358","display_name":"Differential privacy","level":2,"score":0.6751000285148621},{"id":"https://openalex.org/C79974875","wikidata":"https://www.wikidata.org/wiki/Q483639","display_name":"Cloud computing","level":2,"score":0.629800021648407},{"id":"https://openalex.org/C2779231336","wikidata":"https://www.wikidata.org/wiki/Q7534724","display_name":"Sketch","level":2,"score":0.5989000201225281},{"id":"https://openalex.org/C2780165032","wikidata":"https://www.wikidata.org/wiki/Q16869822","display_name":"Energy consumption","level":2,"score":0.5325999855995178},{"id":"https://openalex.org/C2779960059","wikidata":"https://www.wikidata.org/wiki/Q7113681","display_name":"Overhead (engineering)","level":2,"score":0.42590001225471497},{"id":"https://openalex.org/C162307627","wikidata":"https://www.wikidata.org/wiki/Q204833","display_name":"Enhanced Data Rates for GSM Evolution","level":2,"score":0.39980000257492065},{"id":"https://openalex.org/C138236772","wikidata":"https://www.wikidata.org/wiki/Q25098575","display_name":"Edge device","level":3,"score":0.3978999853134155},{"id":"https://openalex.org/C82876162","wikidata":"https://www.wikidata.org/wiki/Q17096504","display_name":"Latency (audio)","level":2,"score":0.36959999799728394},{"id":"https://openalex.org/C123201435","wikidata":"https://www.wikidata.org/wiki/Q456632","display_name":"Information privacy","level":2,"score":0.3662000000476837},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.351500004529953},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.34950000047683716},{"id":"https://openalex.org/C2779439875","wikidata":"https://www.wikidata.org/wiki/Q1078276","display_name":"Natural language understanding","level":3,"score":0.33889999985694885},{"id":"https://openalex.org/C2778456923","wikidata":"https://www.wikidata.org/wiki/Q5337692","display_name":"Edge computing","level":3,"score":0.33869999647140503},{"id":"https://openalex.org/C125411270","wikidata":"https://www.wikidata.org/wiki/Q18653","display_name":"Encoding (memory)","level":2,"score":0.32100000977516174},{"id":"https://openalex.org/C45374587","wikidata":"https://www.wikidata.org/wiki/Q12525525","display_name":"Computation","level":2,"score":0.3073999881744385},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3027999997138977},{"id":"https://openalex.org/C195324797","wikidata":"https://www.wikidata.org/wiki/Q33742","display_name":"Natural language","level":2,"score":0.29649999737739563},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.29190000891685486},{"id":"https://openalex.org/C173853756","wikidata":"https://www.wikidata.org/wiki/Q86915","display_name":"Dialog box","level":2,"score":0.29010000824928284},{"id":"https://openalex.org/C31258907","wikidata":"https://www.wikidata.org/wiki/Q1301371","display_name":"Computer network","level":1,"score":0.2824000120162964},{"id":"https://openalex.org/C184337299","wikidata":"https://www.wikidata.org/wiki/Q1437428","display_name":"Semantics (computer science)","level":2,"score":0.28209999203681946},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.27790001034736633},{"id":"https://openalex.org/C107457646","wikidata":"https://www.wikidata.org/wiki/Q207434","display_name":"Human\u2013computer interaction","level":1,"score":0.27709999680519104},{"id":"https://openalex.org/C46743427","wikidata":"https://www.wikidata.org/wiki/Q1341685","display_name":"Inference engine","level":3,"score":0.2768999934196472},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.2768000066280365},{"id":"https://openalex.org/C45347329","wikidata":"https://www.wikidata.org/wiki/Q5166604","display_name":"Convolution (computer science)","level":3,"score":0.2678999900817871},{"id":"https://openalex.org/C74172769","wikidata":"https://www.wikidata.org/wiki/Q1446839","display_name":"Routing (electronic design automation)","level":2,"score":0.26649999618530273},{"id":"https://openalex.org/C2742236","wikidata":"https://www.wikidata.org/wiki/Q924713","display_name":"Efficient energy use","level":2,"score":0.2565999925136566},{"id":"https://openalex.org/C5119721","wikidata":"https://www.wikidata.org/wiki/Q220501","display_name":"Quality of service","level":2,"score":0.25060001015663147}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1609/aaai.v40i33.40041","is_oa":true,"landing_page_url":"https://doi.org/10.1609/aaai.v40i33.40041","pdf_url":null,"source":{"id":"https://openalex.org/S4210191458","display_name":"Proceedings of the AAAI Conference on Artificial Intelligence","issn_l":"2159-5399","issn":["2159-5399","2374-3468"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/P4310320058","host_organization_name":"Association for the Advancement of Artificial Intelligence","host_organization_lineage":["https://openalex.org/P4310320058"],"host_organization_lineage_names":["Association for the Advancement of Artificial Intelligence"],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the AAAI Conference on Artificial Intelligence","raw_type":"journal-article"}],"best_oa_location":{"id":"doi:10.1609/aaai.v40i33.40041","is_oa":true,"landing_page_url":"https://doi.org/10.1609/aaai.v40i33.40041","pdf_url":null,"source":{"id":"https://openalex.org/S4210191458","display_name":"Proceedings of the AAAI Conference on Artificial Intelligence","issn_l":"2159-5399","issn":["2159-5399","2374-3468"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/P4310320058","host_organization_name":"Association for the Advancement of Artificial Intelligence","host_organization_lineage":["https://openalex.org/P4310320058"],"host_organization_lineage_names":["Association for the Advancement of Artificial Intelligence"],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the AAAI Conference on Artificial Intelligence","raw_type":"journal-article"},"sustainable_development_goals":[{"score":0.8930056095123291,"display_name":"Affordable and clean energy","id":"https://metadata.un.org/sdg/7"}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Large":[0],"Language":[1],"Models":[2],"(LLMs)":[3],"demonstrate":[4],"impressive":[5],"capabilities":[6],"in":[7,21,45,68,104,183],"natural":[8],"language":[9,168],"understanding":[10],"and":[11,18,27,71,99,147,189,199,224],"generation,":[12],"but":[13],"incur":[14],"high":[15,204],"communication":[16],"overhead":[17],"privacy":[19,44,61,98,142,209],"risks":[20],"cloud":[22,150],"deployments,":[23],"while":[24,202],"facing":[25],"compute":[26],"memory":[28],"constraints":[29],"when":[30],"confined":[31],"to":[32,191],"edge":[33,109,136],"devices.Cloud\u2013edge":[34],"inference":[35,57,100],"has":[36],"emerged":[37],"as":[38,197],"a":[39,92,115,153],"promising":[40],"paradigm":[41],"for":[42,75,86,132],"improving":[43],"LLM":[46,151],"services":[47],"by":[48,164],"retaining":[49],"sensitive":[50],"computations":[51],"on":[52,120,144],"local":[53,140,172],"devices.However,":[54],"existing":[55],"cloud\u2013edge":[56,226],"approaches":[58],"apply":[59],"uniform":[60],"protection":[62],"without":[63],"considering":[64],"input":[65],"sensitivity,":[66],"resulting":[67],"unnecessary":[69],"perturbation":[70],"degraded":[72],"utility":[73],"even":[74],"non-sensitive":[76],"tokens.":[77],"To":[78],"address":[79],"this":[80],"limitation,":[81],"we":[82],"propose":[83],"Privacy-aware":[84],"Routing":[85],"Inference":[87],"with":[88],"Semantic":[89],"Modulation":[90],"(PRISM),":[91],"context-aware":[93],"framework":[94],"that":[95,176],"dynamically":[96],"balances":[97],"quality.":[101],"PRISM":[102,177],"executes":[103],"four":[105],"stages:":[106],"(1)":[107],"the":[108,121,135,149,157,165],"device":[110],"profiles":[111],"entity-level":[112],"sensitivity;":[113],"(2)":[114],"soft":[116],"gating":[117],"module,":[118],"also":[119],"edge,":[122,128],"selects":[123],"an":[124],"execution":[125],"mode":[126],"-cloud,":[127],"or":[129],"collaboration;":[130],"(3)":[131],"collaborative":[133],"paths,":[134],"applies":[137],"adaptive":[138],"two-layer":[139],"differential":[141],"based":[143],"entity":[145],"risks;":[146],"(4)":[148],"generates":[152],"semantic":[154],"sketch":[155],"from":[156],"perturbed":[158],"prompt,":[159],"which":[160],"is":[161],"then":[162],"refined":[163],"edge-side":[166],"small":[167],"model":[169,227],"(SLM)":[170],"using":[171],"context.Our":[173],"results":[174],"show":[175],"consistently":[178],"achieves":[179],"superior":[180],"privacy-utility":[181],"trade-offs":[182],"various":[184],"scenarios,":[185],"reducing":[186],"energy":[187,222],"consumption":[188],"latency":[190],"40\u201350%":[192],"of":[193],"baseline":[194],"methods":[195],"such":[196],"Uniform":[198],"Selective":[200],"LDP,":[201],"maintaining":[203],"output":[205],"quality":[206],"under":[207],"strong":[208],"constraints.":[210],"These":[211],"findings":[212],"are":[213],"validated":[214],"through":[215],"comprehensive":[216],"evaluations":[217],"involving":[218],"realistic":[219],"prompts,":[220],"actual":[221],"measurements,":[223],"heterogeneous":[225],"deployments.":[228]},"counts_by_year":[{"year":2026,"cited_by_count":1}],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2026-03-18T00:00:00"}