{"id":"https://openalex.org/W7106316903","doi":"https://doi.org/10.1145/3680207.3765696","title":"Poster: MoE <sup>2</sup> : Optimizing Collaborative Inference for Edge Large Language Models","display_name":"Poster: MoE <sup>2</sup> : Optimizing Collaborative Inference for Edge Large Language Models","publication_year":2025,"publication_date":"2025-11-03","ids":{"openalex":"https://openalex.org/W7106316903","doi":"https://doi.org/10.1145/3680207.3765696"},"language":null,"primary_location":{"id":"doi:10.1145/3680207.3765696","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3680207.3765696","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3680207.3765696","source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 31st Annual International Conference on Mobile Computing and Networking","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/10.1145/3680207.3765696","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":null,"display_name":"Lyudong Jin","orcid":"https://orcid.org/0009-0003-2454-6577"},"institutions":[{"id":"https://openalex.org/I4210158570","display_name":"Zhejiang University-University of Edinburgh Institute","ror":"https://ror.org/04jth1r26","country_code":"CN","type":"education","lineage":["https://openalex.org/I4210158570","https://openalex.org/I76130692","https://openalex.org/I98677209"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Lyudong Jin","raw_affiliation_strings":["ZJU-UIUC Institute, Zhejiang University, Haining, China"],"raw_orcid":"https://orcid.org/0009-0003-2454-6577","affiliations":[{"raw_affiliation_string":"ZJU-UIUC Institute, Zhejiang University, Haining, China","institution_ids":["https://openalex.org/I4210158570"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Yanning Zhang","orcid":"https://orcid.org/0009-0009-0111-2761"},"institutions":[{"id":"https://openalex.org/I4210158570","display_name":"Zhejiang University-University of Edinburgh Institute","ror":"https://ror.org/04jth1r26","country_code":"CN","type":"education","lineage":["https://openalex.org/I4210158570","https://openalex.org/I76130692","https://openalex.org/I98677209"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yanning Zhang","raw_affiliation_strings":["ZJU-UIUC Institute, Zhejiang University, Haining, China"],"raw_orcid":"https://orcid.org/0009-0009-0111-2761","affiliations":[{"raw_affiliation_string":"ZJU-UIUC Institute, Zhejiang University, Haining, China","institution_ids":["https://openalex.org/I4210158570"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Yanhan Li","orcid":"https://orcid.org/0009-0003-6291-8132"},"institutions":[{"id":"https://openalex.org/I4210158570","display_name":"Zhejiang University-University of Edinburgh Institute","ror":"https://ror.org/04jth1r26","country_code":"CN","type":"education","lineage":["https://openalex.org/I4210158570","https://openalex.org/I76130692","https://openalex.org/I98677209"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yanhan Li","raw_affiliation_strings":["ZJU-UIUC Institute, Zhejiang University, Haining, China"],"raw_orcid":"https://orcid.org/0009-0003-6291-8132","affiliations":[{"raw_affiliation_string":"ZJU-UIUC Institute, Zhejiang University, Haining, China","institution_ids":["https://openalex.org/I4210158570"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Shurong Wang","orcid":"https://orcid.org/0009-0004-5221-5043"},"institutions":[{"id":"https://openalex.org/I4210158570","display_name":"Zhejiang University-University of Edinburgh Institute","ror":"https://ror.org/04jth1r26","country_code":"CN","type":"education","lineage":["https://openalex.org/I4210158570","https://openalex.org/I76130692","https://openalex.org/I98677209"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Shurong Wang","raw_affiliation_strings":["ZJU-UIUC Institute, Zhejiang University, Haining, China"],"raw_orcid":"https://orcid.org/0009-0004-5221-5043","affiliations":[{"raw_affiliation_string":"ZJU-UIUC Institute, Zhejiang University, Haining, China","institution_ids":["https://openalex.org/I4210158570"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Howard H. Yang","orcid":"https://orcid.org/0000-0002-0256-2416"},"institutions":[{"id":"https://openalex.org/I4210158570","display_name":"Zhejiang University-University of Edinburgh Institute","ror":"https://ror.org/04jth1r26","country_code":"CN","type":"education","lineage":["https://openalex.org/I4210158570","https://openalex.org/I76130692","https://openalex.org/I98677209"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Howard H. Yang","raw_affiliation_strings":["ZJU-UIUC Institute, Zhejiang University, Haining, China"],"raw_orcid":"https://orcid.org/0000-0002-0256-2416","affiliations":[{"raw_affiliation_string":"ZJU-UIUC Institute, Zhejiang University, Haining, China","institution_ids":["https://openalex.org/I4210158570"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Jian Wu","orcid":"https://orcid.org/0000-0002-3230-6392"},"institutions":[{"id":"https://openalex.org/I76130692","display_name":"Zhejiang University","ror":"https://ror.org/00a2xv884","country_code":"CN","type":"education","lineage":["https://openalex.org/I76130692"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Jian Wu","raw_affiliation_strings":["Zhejiang University, Hangzhou, China"],"raw_orcid":"https://orcid.org/0000-0002-3230-6392","affiliations":[{"raw_affiliation_string":"Zhejiang University, Hangzhou, China","institution_ids":["https://openalex.org/I76130692"]}]},{"author_position":"last","author":{"id":null,"display_name":"Meng Zhang","orcid":"https://orcid.org/0000-0002-4893-6946"},"institutions":[{"id":"https://openalex.org/I4210158570","display_name":"Zhejiang University-University of Edinburgh Institute","ror":"https://ror.org/04jth1r26","country_code":"CN","type":"education","lineage":["https://openalex.org/I4210158570","https://openalex.org/I76130692","https://openalex.org/I98677209"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Meng Zhang","raw_affiliation_strings":["ZJU-UIUC Institute, Zhejiang University, Haining, China"],"raw_orcid":"https://orcid.org/0000-0002-4893-6946","affiliations":[{"raw_affiliation_string":"ZJU-UIUC Institute, Zhejiang University, Haining, China","institution_ids":["https://openalex.org/I4210158570"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":5.019,"has_fulltext":true,"cited_by_count":3,"citation_normalized_percentile":{"value":0.96139875,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":94,"max":97},"biblio":{"volume":null,"issue":null,"first_page":"1389","last_page":"1391"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.1915999948978424,"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":0.1915999948978424,"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/T11714","display_name":"Multimodal Machine Learning Applications","score":0.07819999754428864,"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/T11636","display_name":"Artificial Intelligence in Healthcare and Education","score":0.06840000301599503,"subfield":{"id":"https://openalex.org/subfields/2718","display_name":"Health Informatics"},"field":{"id":"https://openalex.org/fields/27","display_name":"Medicine"},"domain":{"id":"https://openalex.org/domains/4","display_name":"Health Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.7466999888420105},{"id":"https://openalex.org/keywords/leverage","display_name":"Leverage (statistics)","score":0.7305999994277954},{"id":"https://openalex.org/keywords/enhanced-data-rates-for-gsm-evolution","display_name":"Enhanced Data Rates for GSM Evolution","score":0.5647000074386597},{"id":"https://openalex.org/keywords/selection","display_name":"Selection (genetic algorithm)","score":0.5198000073432922},{"id":"https://openalex.org/keywords/monotonic-function","display_name":"Monotonic function","score":0.46540001034736633},{"id":"https://openalex.org/keywords/property","display_name":"Property (philosophy)","score":0.4341000020503998},{"id":"https://openalex.org/keywords/exploit","display_name":"Exploit","score":0.387800008058548},{"id":"https://openalex.org/keywords/edge-device","display_name":"Edge device","score":0.382999986410141}],"concepts":[{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.7466999888420105},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7389000058174133},{"id":"https://openalex.org/C153083717","wikidata":"https://www.wikidata.org/wiki/Q6535263","display_name":"Leverage (statistics)","level":2,"score":0.7305999994277954},{"id":"https://openalex.org/C162307627","wikidata":"https://www.wikidata.org/wiki/Q204833","display_name":"Enhanced Data Rates for GSM Evolution","level":2,"score":0.5647000074386597},{"id":"https://openalex.org/C81917197","wikidata":"https://www.wikidata.org/wiki/Q628760","display_name":"Selection (genetic algorithm)","level":2,"score":0.5198000073432922},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4819999933242798},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.47679999470710754},{"id":"https://openalex.org/C72169020","wikidata":"https://www.wikidata.org/wiki/Q194404","display_name":"Monotonic function","level":2,"score":0.46540001034736633},{"id":"https://openalex.org/C189950617","wikidata":"https://www.wikidata.org/wiki/Q937228","display_name":"Property (philosophy)","level":2,"score":0.4341000020503998},{"id":"https://openalex.org/C165696696","wikidata":"https://www.wikidata.org/wiki/Q11287","display_name":"Exploit","level":2,"score":0.387800008058548},{"id":"https://openalex.org/C138236772","wikidata":"https://www.wikidata.org/wiki/Q25098575","display_name":"Edge device","level":3,"score":0.382999986410141},{"id":"https://openalex.org/C46743427","wikidata":"https://www.wikidata.org/wiki/Q1341685","display_name":"Inference engine","level":3,"score":0.3359000086784363},{"id":"https://openalex.org/C58328972","wikidata":"https://www.wikidata.org/wiki/Q184609","display_name":"Expert system","level":2,"score":0.3301999866962433},{"id":"https://openalex.org/C55166926","wikidata":"https://www.wikidata.org/wiki/Q2892946","display_name":"Oracle","level":2,"score":0.313400000333786},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.29760000109672546},{"id":"https://openalex.org/C137836250","wikidata":"https://www.wikidata.org/wiki/Q984063","display_name":"Optimization problem","level":2,"score":0.29600000381469727},{"id":"https://openalex.org/C82876162","wikidata":"https://www.wikidata.org/wiki/Q17096504","display_name":"Latency (audio)","level":2,"score":0.2915000021457672},{"id":"https://openalex.org/C206345919","wikidata":"https://www.wikidata.org/wiki/Q20380951","display_name":"Resource (disambiguation)","level":2,"score":0.27959999442100525},{"id":"https://openalex.org/C51823790","wikidata":"https://www.wikidata.org/wiki/Q504353","display_name":"Greedy algorithm","level":2,"score":0.27480000257492065},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.27399998903274536},{"id":"https://openalex.org/C195324797","wikidata":"https://www.wikidata.org/wiki/Q33742","display_name":"Natural language","level":2,"score":0.2720000147819519},{"id":"https://openalex.org/C2777472644","wikidata":"https://www.wikidata.org/wiki/Q16968992","display_name":"Approximate inference","level":3,"score":0.2718999981880188},{"id":"https://openalex.org/C22367795","wikidata":"https://www.wikidata.org/wiki/Q7625208","display_name":"Structured prediction","level":2,"score":0.26750001311302185},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.2669000029563904}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3680207.3765696","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3680207.3765696","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3680207.3765696","source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 31st Annual International Conference on Mobile Computing and Networking","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.1145/3680207.3765696","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3680207.3765696","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3680207.3765696","source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 31st Annual International Conference on Mobile Computing and Networking","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G2823080268","display_name":null,"funder_award_id":"62572433","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"}],"funders":[{"id":"https://openalex.org/F4320321001","display_name":"National Natural Science Foundation of China","ror":"https://ror.org/01h0zpd94"}],"has_content":{"grobid_xml":false,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7106316903.pdf"},"referenced_works_count":1,"referenced_works":["https://openalex.org/W4406548238"],"related_works":[],"abstract_inverted_index":{"Large":[0],"language":[1,11],"models":[2],"(LLMs)":[3],"have":[4],"demonstrated":[5],"remarkable":[6],"capabilities":[7,17],"across":[8,133],"various":[9,141],"natural":[10],"processing":[12],"tasks.":[13],"Exploiting":[14],"the":[15,48,75,111],"heterogeneous":[16],"of":[18,80,98],"edge":[19,44,81],"LLMs":[20],"is":[21,70],"crucial":[22],"for":[23,43],"emerging":[24],"applications,":[25],"enabling":[26],"greater":[27],"cost-effectiveness":[28],"and":[29,51,60,78,93,104,114,125,135],"reduced":[30],"latency.":[31],"In":[32],"this":[33],"work,":[34],"we":[35,86],"introduce":[36],"Mixture-of-Edge-Experts":[37],"(MoE2),":[38],"a":[39,88,116],"collaborative":[40],"inference":[41,57],"framework":[42],"LLMs.":[45,82],"We":[46,108],"formulate":[47],"joint":[49],"gating":[50,99],"expert":[52,67,90],"selection":[53,68,91],"problem":[54],"to":[55,74],"optimize":[56],"under":[58,140],"energy":[59,136],"latency":[61],"constraints.":[62,143],"Unlike":[63],"conventional":[64],"MoE":[65],"problems,":[66],"here":[69],"more":[71],"challenging":[72],"due":[73],"combinatorial":[76],"nature":[77],"heterogeneity":[79],"To":[83],"address":[84],"this,":[85],"propose":[87],"two-level":[89],"mechanism":[92],"uncover":[94],"an":[95],"optimality-preserving":[96],"property":[97],"parameters":[100],"that":[101],"decouples":[102],"training":[103],"selection,":[105],"reducing":[106],"complexity.":[107],"further":[109],"leverage":[110],"objective's":[112],"monotonicity":[113],"design":[115],"discrete":[117],"monotonic":[118],"optimization":[119],"algorithm.":[120],"Implemented":[121],"on":[122],"Jetson":[123],"Orin":[124],"RTX":[126],"4090":[127],"platforms,":[128],"MoE2":[129],"achieves":[130],"optimal":[131],"trade-offs":[132],"delay":[134],"budgets,":[137],"outperforming":[138],"baselines":[139],"resource":[142]},"counts_by_year":[{"year":2026,"cited_by_count":1},{"year":2025,"cited_by_count":2}],"updated_date":"2026-07-29T14:22:42.915294","created_date":"2025-11-23T00:00:00"}