{"id":"https://openalex.org/W4410228170","doi":"https://doi.org/10.1109/wcnc61545.2025.10978177","title":"Edge vs Cloud: How Do We Balance Cost, Latency, and Quality for Large Language Models Over 5G Networks?","display_name":"Edge vs Cloud: How Do We Balance Cost, Latency, and Quality for Large Language Models Over 5G Networks?","publication_year":2025,"publication_date":"2025-03-24","ids":{"openalex":"https://openalex.org/W4410228170","doi":"https://doi.org/10.1109/wcnc61545.2025.10978177"},"language":"en","primary_location":{"id":"doi:10.1109/wcnc61545.2025.10978177","is_oa":false,"landing_page_url":"https://doi.org/10.1109/wcnc61545.2025.10978177","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 IEEE Wireless Communications and Networking Conference (WCNC)","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":false,"oa_status":"closed","oa_url":null,"any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5100343632","display_name":"Minsu Kim","orcid":"https://orcid.org/0000-0002-7556-8958"},"institutions":[{"id":"https://openalex.org/I859038795","display_name":"Virginia Tech","ror":"https://ror.org/02smfhw86","country_code":"US","type":"education","lineage":["https://openalex.org/I859038795"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Minsu Kim","raw_affiliation_strings":["Virginia Tech,Bradley Department of Electrical and Computer Engineering,Arlington,VA,USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Virginia Tech,Bradley Department of Electrical and Computer Engineering,Arlington,VA,USA","institution_ids":["https://openalex.org/I859038795"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5064810159","display_name":"Pinyarash Pinyoanuntapong","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Pinyarash Pinyoanuntapong","raw_affiliation_strings":["MediaTek,Warren,NJ,USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"MediaTek,Warren,NJ,USA","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5048604096","display_name":"Bongho Kim","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Bongho Kim","raw_affiliation_strings":["MediaTek,Warren,NJ,USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"MediaTek,Warren,NJ,USA","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5108117710","display_name":"Walid Saad","orcid":null},"institutions":[{"id":"https://openalex.org/I859038795","display_name":"Virginia Tech","ror":"https://ror.org/02smfhw86","country_code":"US","type":"education","lineage":["https://openalex.org/I859038795"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Walid Saad","raw_affiliation_strings":["Virginia Tech,Bradley Department of Electrical and Computer Engineering,Arlington,VA,USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Virginia Tech,Bradley Department of Electrical and Computer Engineering,Arlington,VA,USA","institution_ids":["https://openalex.org/I859038795"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5110512574","display_name":"Doru Calin","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Doru Calin","raw_affiliation_strings":["MediaTek,Warren,NJ,USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"MediaTek,Warren,NJ,USA","institution_ids":[]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":5,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":7.0351,"has_fulltext":false,"cited_by_count":4,"citation_normalized_percentile":{"value":0.96437946,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":91,"max":99},"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"6"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.8472999930381775,"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/T10028","display_name":"Topic Modeling","score":0.8472999930381775,"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/T13382","display_name":"Robotics and Automated Systems","score":0.82669997215271,"subfield":{"id":"https://openalex.org/subfields/2207","display_name":"Control and Systems 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/T10203","display_name":"Recommender Systems and Techniques","score":0.8130000233650208,"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/computer-science","display_name":"Computer science","score":0.7763183116912842},{"id":"https://openalex.org/keywords/latency","display_name":"Latency (audio)","score":0.6760938167572021},{"id":"https://openalex.org/keywords/cloud-computing","display_name":"Cloud computing","score":0.6391274333000183},{"id":"https://openalex.org/keywords/enhanced-data-rates-for-gsm-evolution","display_name":"Enhanced Data Rates for GSM Evolution","score":0.5686637759208679},{"id":"https://openalex.org/keywords/balance","display_name":"Balance (ability)","score":0.5076166391372681},{"id":"https://openalex.org/keywords/computer-network","display_name":"Computer network","score":0.3904837965965271},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.15342861413955688},{"id":"https://openalex.org/keywords/operating-system","display_name":"Operating system","score":0.13247522711753845},{"id":"https://openalex.org/keywords/telecommunications","display_name":"Telecommunications","score":0.11215385794639587}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7763183116912842},{"id":"https://openalex.org/C82876162","wikidata":"https://www.wikidata.org/wiki/Q17096504","display_name":"Latency (audio)","level":2,"score":0.6760938167572021},{"id":"https://openalex.org/C79974875","wikidata":"https://www.wikidata.org/wiki/Q483639","display_name":"Cloud computing","level":2,"score":0.6391274333000183},{"id":"https://openalex.org/C162307627","wikidata":"https://www.wikidata.org/wiki/Q204833","display_name":"Enhanced Data Rates for GSM Evolution","level":2,"score":0.5686637759208679},{"id":"https://openalex.org/C168031717","wikidata":"https://www.wikidata.org/wiki/Q1530280","display_name":"Balance (ability)","level":2,"score":0.5076166391372681},{"id":"https://openalex.org/C31258907","wikidata":"https://www.wikidata.org/wiki/Q1301371","display_name":"Computer network","level":1,"score":0.3904837965965271},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.15342861413955688},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.13247522711753845},{"id":"https://openalex.org/C76155785","wikidata":"https://www.wikidata.org/wiki/Q418","display_name":"Telecommunications","level":1,"score":0.11215385794639587},{"id":"https://openalex.org/C99508421","wikidata":"https://www.wikidata.org/wiki/Q2678675","display_name":"Physical medicine and rehabilitation","level":1,"score":0.0},{"id":"https://openalex.org/C71924100","wikidata":"https://www.wikidata.org/wiki/Q11190","display_name":"Medicine","level":0,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/wcnc61545.2025.10978177","is_oa":false,"landing_page_url":"https://doi.org/10.1109/wcnc61545.2025.10978177","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 IEEE Wireless Communications and Networking Conference (WCNC)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[{"score":0.4399999976158142,"id":"https://metadata.un.org/sdg/4","display_name":"Quality Education"}],"awards":[{"id":"https://openalex.org/G6971433196","display_name":null,"funder_award_id":"CNS-2114267","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"}],"funders":[{"id":"https://openalex.org/F4320306076","display_name":"National Science Foundation","ror":"https://ror.org/021nxhr62"}],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":13,"referenced_works":["https://openalex.org/W2962855623","https://openalex.org/W3046577719","https://openalex.org/W3179971958","https://openalex.org/W3196647046","https://openalex.org/W4383751223","https://openalex.org/W4385571189","https://openalex.org/W4390513007","https://openalex.org/W4392026544","https://openalex.org/W4402202526","https://openalex.org/W4408223168","https://openalex.org/W6797816506","https://openalex.org/W6852874933","https://openalex.org/W6865013197"],"related_works":["https://openalex.org/W4244478748","https://openalex.org/W3150465815","https://openalex.org/W4223488648","https://openalex.org/W2134969820","https://openalex.org/W2251605416","https://openalex.org/W1997222214","https://openalex.org/W2560439919","https://openalex.org/W4389340727","https://openalex.org/W2802581102","https://openalex.org/W4205786897"],"abstract_inverted_index":{"Large":[0],"language":[1],"models":[2],"(LLMs)":[3],"can":[4,30,196],"perform":[5],"a":[6,52,59,81,94,101,109,157,175,209,215],"plethora":[7],"of":[8,72,129,148,165],"tasks,":[9],"however,":[10],"they":[11,40],"often":[12,41],"require":[13],"cloud":[14],"servers":[15],"for":[16,58],"deployment":[17],"due":[18],"to":[19,93,116,122,214],"their":[20],"computing":[21],"cost":[22,199],"and":[23,27,75,127,146,152,200,204],"size.":[24],"Meanwhile,":[25],"small":[26],"cost-effective":[28],"LLMs":[29],"be":[31],"deployed":[32],"on":[33,100,156,184],"edge":[34],"devices":[35],"(e.g.,":[36],"mobile":[37,82,106,158],"devices),":[38],"but":[39],"exhibit":[42],"lower":[43],"response":[44],"quality":[45],"than":[46],"larger":[47],"models.":[48],"In":[49,77],"this":[50],"paper,":[51],"measurement-driven":[53],"training":[54],"framework":[55,195],"is":[56,139,169,180],"proposed":[57,182,194],"device-side":[60],"artificial":[61],"intelligence":[62],"(AI)-enabled":[63],"router":[64,110],"that":[65,111,192,217],"selects":[66],"the":[67,78,117,124,135,137,143,149,153,166,185,193,198,220],"best":[68],"LLM":[69,87,97,121,131,151,155],"in":[70],"terms":[71],"cost,":[73,125,144],"latency,":[74,126,145],"performance.":[76],"considered":[79],"framework,":[80],"device":[83,107],"uses":[84,219],"its":[85],"local":[86,118,154],"(sub-billion":[88],"LLM),":[89],"while":[90],"having":[91],"access":[92],"bigger":[95],"server":[96,120,150,221],"(GPT-4)":[98],"hosted":[99],"5G":[102],"core":[103],"network.":[104],"The":[105,160],"has":[108],"sends":[112],"an":[113],"input":[114],"prompt":[115],"or":[119],"optimize":[123],"performance":[128,147,211],"output":[130],"responses.":[132],"To":[133,171],"train":[134],"router,":[136],"dataset":[138,168],"constructed":[140],"by":[141,202],"measuring":[142],"device.":[159],"structural":[161],"causal":[162],"model":[163],"(SCM)":[164],"measured":[167],"identified.":[170],"further":[172],"improve":[173,197],"performance,":[174],"causality-driven":[176],"data":[177],"augmentation":[178],"method":[179],"also":[181],"based":[183],"discovered":[186],"SCM.":[187],"Real-world":[188],"experimental":[189],"results":[190],"show":[191],"latency":[201],"50%":[203],"31%,":[205],"respectively,":[206],"with":[207],"only":[208,218],"2.13%":[210],"loss":[212],"compared":[213],"baseline":[216],"LLM.":[222]},"counts_by_year":[{"year":2026,"cited_by_count":3},{"year":2025,"cited_by_count":1}],"updated_date":"2026-06-11T09:08:48.828518","created_date":"2025-10-10T00:00:00"}