{"id":"https://openalex.org/W7125980547","doi":"https://doi.org/10.1109/ton.2026.3658387","title":"Resource-Efficient LLM Customization on Mobile Devices Through Proxy Submodel Tuning","display_name":"Resource-Efficient LLM Customization on Mobile Devices Through Proxy Submodel Tuning","publication_year":2026,"publication_date":"2026-01-01","ids":{"openalex":"https://openalex.org/W7125980547","doi":"https://doi.org/10.1109/ton.2026.3658387"},"language":null,"primary_location":{"id":"doi:10.1109/ton.2026.3658387","is_oa":false,"landing_page_url":"https://doi.org/10.1109/ton.2026.3658387","pdf_url":null,"source":{"id":"https://openalex.org/S5407042750","display_name":"IEEE Transactions on Networking","issn_l":"2998-4157","issn":["2998-4157"],"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Transactions on Networking","raw_type":"journal-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/A5124072491","display_name":"Yan Zhuang","orcid":null},"institutions":[{"id":"https://openalex.org/I183067930","display_name":"Shanghai Jiao Tong University","ror":"https://ror.org/0220qvk04","country_code":"CN","type":"education","lineage":["https://openalex.org/I183067930"]}],"countries":["CN"],"is_corresponding":true,"raw_author_name":"Yan Zhuang","raw_affiliation_strings":["Department of Computer Science and Engineering, Shanghai Key Laboratory of Scalable Computing and Systems, Shanghai Jiao Tong University, Shanghai, China"],"raw_orcid":"https://orcid.org/0009-0006-0702-5093","affiliations":[{"raw_affiliation_string":"Department of Computer Science and Engineering, Shanghai Key Laboratory of Scalable Computing and Systems, Shanghai Jiao Tong University, Shanghai, China","institution_ids":["https://openalex.org/I183067930"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5124130755","display_name":"Chen Gong","orcid":null},"institutions":[{"id":"https://openalex.org/I183067930","display_name":"Shanghai Jiao Tong University","ror":"https://ror.org/0220qvk04","country_code":"CN","type":"education","lineage":["https://openalex.org/I183067930"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Chen Gong","raw_affiliation_strings":["Department of Computer Science and Engineering, Shanghai Key Laboratory of Scalable Computing and Systems, Shanghai Jiao Tong University, Shanghai, China"],"raw_orcid":"https://orcid.org/0000-0003-0333-6418","affiliations":[{"raw_affiliation_string":"Department of Computer Science and Engineering, Shanghai Key Laboratory of Scalable Computing and Systems, Shanghai Jiao Tong University, Shanghai, China","institution_ids":["https://openalex.org/I183067930"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5124108493","display_name":"Zhenzhe Zheng","orcid":null},"institutions":[{"id":"https://openalex.org/I183067930","display_name":"Shanghai Jiao Tong University","ror":"https://ror.org/0220qvk04","country_code":"CN","type":"education","lineage":["https://openalex.org/I183067930"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zhenzhe Zheng","raw_affiliation_strings":["Department of Computer Science and Engineering, Shanghai Key Laboratory of Scalable Computing and Systems, Shanghai Jiao Tong University, Shanghai, China"],"raw_orcid":"https://orcid.org/0000-0003-3447-5349","affiliations":[{"raw_affiliation_string":"Department of Computer Science and Engineering, Shanghai Key Laboratory of Scalable Computing and Systems, Shanghai Jiao Tong University, Shanghai, China","institution_ids":["https://openalex.org/I183067930"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5017118639","display_name":"F. Y. Wu","orcid":null},"institutions":[{"id":"https://openalex.org/I183067930","display_name":"Shanghai Jiao Tong University","ror":"https://ror.org/0220qvk04","country_code":"CN","type":"education","lineage":["https://openalex.org/I183067930"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Fan Wu","raw_affiliation_strings":["Department of Computer Science and Engineering, Shanghai Key Laboratory of Scalable Computing and Systems, Shanghai Jiao Tong University, Shanghai, China"],"raw_orcid":"https://orcid.org/0000-0003-0965-9058","affiliations":[{"raw_affiliation_string":"Department of Computer Science and Engineering, Shanghai Key Laboratory of Scalable Computing and Systems, Shanghai Jiao Tong University, Shanghai, China","institution_ids":["https://openalex.org/I183067930"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5121620022","display_name":"Guihai Chen","orcid":null},"institutions":[{"id":"https://openalex.org/I183067930","display_name":"Shanghai Jiao Tong University","ror":"https://ror.org/0220qvk04","country_code":"CN","type":"education","lineage":["https://openalex.org/I183067930"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Guihai Chen","raw_affiliation_strings":["Department of Computer Science and Engineering, Shanghai Key Laboratory of Scalable Computing and Systems, Shanghai Jiao Tong University, Shanghai, China"],"raw_orcid":"https://orcid.org/0000-0002-6934-1685","affiliations":[{"raw_affiliation_string":"Department of Computer Science and Engineering, Shanghai Key Laboratory of Scalable Computing and Systems, Shanghai Jiao Tong University, Shanghai, China","institution_ids":["https://openalex.org/I183067930"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":5,"corresponding_author_ids":["https://openalex.org/A5124072491"],"corresponding_institution_ids":["https://openalex.org/I183067930"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.15677866,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":"34","issue":null,"first_page":"3687","last_page":"3703"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T12127","display_name":"Software System Performance and Reliability","score":0.27619999647140503,"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"}},"topics":[{"id":"https://openalex.org/T12127","display_name":"Software System Performance and Reliability","score":0.27619999647140503,"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/T11241","display_name":"Advanced Malware Detection Techniques","score":0.14710000157356262,"subfield":{"id":"https://openalex.org/subfields/1711","display_name":"Signal Processing"},"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/T10260","display_name":"Software Engineering Research","score":0.09290000051259995,"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/mobile-device","display_name":"Mobile device","score":0.7641000151634216},{"id":"https://openalex.org/keywords/scalability","display_name":"Scalability","score":0.704800009727478},{"id":"https://openalex.org/keywords/merge","display_name":"Merge (version control)","score":0.6171000003814697},{"id":"https://openalex.org/keywords/personalization","display_name":"Personalization","score":0.5206999778747559},{"id":"https://openalex.org/keywords/reuse","display_name":"Reuse","score":0.43130001425743103},{"id":"https://openalex.org/keywords/mobile-computing","display_name":"Mobile computing","score":0.4081999957561493},{"id":"https://openalex.org/keywords/proxy","display_name":"Proxy (statistics)","score":0.3684000074863434}],"concepts":[{"id":"https://openalex.org/C186967261","wikidata":"https://www.wikidata.org/wiki/Q5082128","display_name":"Mobile device","level":2,"score":0.7641000151634216},{"id":"https://openalex.org/C48044578","wikidata":"https://www.wikidata.org/wiki/Q727490","display_name":"Scalability","level":2,"score":0.704800009727478},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6948999762535095},{"id":"https://openalex.org/C197129107","wikidata":"https://www.wikidata.org/wiki/Q1921621","display_name":"Merge (version control)","level":2,"score":0.6171000003814697},{"id":"https://openalex.org/C183003079","wikidata":"https://www.wikidata.org/wiki/Q1000371","display_name":"Personalization","level":2,"score":0.5206999778747559},{"id":"https://openalex.org/C206588197","wikidata":"https://www.wikidata.org/wiki/Q846574","display_name":"Reuse","level":2,"score":0.43130001425743103},{"id":"https://openalex.org/C144543869","wikidata":"https://www.wikidata.org/wiki/Q2738570","display_name":"Mobile computing","level":2,"score":0.4081999957561493},{"id":"https://openalex.org/C2780148112","wikidata":"https://www.wikidata.org/wiki/Q1432581","display_name":"Proxy (statistics)","level":2,"score":0.3684000074863434},{"id":"https://openalex.org/C177284502","wikidata":"https://www.wikidata.org/wiki/Q1005390","display_name":"Adapter (computing)","level":2,"score":0.3538999855518341},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.3528999984264374},{"id":"https://openalex.org/C95491727","wikidata":"https://www.wikidata.org/wiki/Q992968","display_name":"Mobile telephony","level":3,"score":0.34619998931884766},{"id":"https://openalex.org/C2779628075","wikidata":"https://www.wikidata.org/wiki/Q1253258","display_name":"Downgrade","level":2,"score":0.3181999921798706},{"id":"https://openalex.org/C60952562","wikidata":"https://www.wikidata.org/wiki/Q6887246","display_name":"Mobile technology","level":3,"score":0.3068000078201294},{"id":"https://openalex.org/C149635348","wikidata":"https://www.wikidata.org/wiki/Q193040","display_name":"Embedded system","level":1,"score":0.3003000020980835},{"id":"https://openalex.org/C557433098","wikidata":"https://www.wikidata.org/wiki/Q94","display_name":"Android (operating system)","level":2,"score":0.27300000190734863},{"id":"https://openalex.org/C31258907","wikidata":"https://www.wikidata.org/wiki/Q1301371","display_name":"Computer network","level":1,"score":0.2621000111103058}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/ton.2026.3658387","is_oa":false,"landing_page_url":"https://doi.org/10.1109/ton.2026.3658387","pdf_url":null,"source":{"id":"https://openalex.org/S5407042750","display_name":"IEEE Transactions on Networking","issn_l":"2998-4157","issn":["2998-4157"],"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Transactions on Networking","raw_type":"journal-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G2436440598","display_name":null,"funder_award_id":"62322206","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"},{"id":"https://openalex.org/G6028208134","display_name":null,"funder_award_id":"2023YFB4502400","funder_id":"https://openalex.org/F4320335777","funder_display_name":"National Key Research and Development Program of China"}],"funders":[{"id":"https://openalex.org/F4320321001","display_name":"National Natural Science Foundation of China","ror":"https://ror.org/01h0zpd94"},{"id":"https://openalex.org/F4320335777","display_name":"National Key Research and Development Program of China","ror":null}],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Considering":[0],"limited":[1,56],"on-device":[2,57],"resources,":[3],"current":[4],"practices":[5],"are":[6,31,190],"attempting":[7],"to":[8,20,33,53,90,127,157,192],"deploy":[9],"a":[10,17,80,106,129,140,167],"system-level":[11],"mixture-of-experts":[12],"(MoE)-based":[13],"foundation":[14,68,137,212],"LLM":[15,44,138],"on":[16,95,196],"mobile":[18,26,29,88,179,188],"device":[19],"serve":[21],"multiple":[22],"apps":[23,30,64,89,189],"and":[24,59,118,123,131,147,153,161,171,178,203],"support":[25],"intelligence.":[27],"However,":[28],"hard":[32],"customize":[34],"their":[35],"services":[36],"that":[37,63,86,185],"require":[38],"fine-tuning":[39],"adapters":[40,94,195],"associated":[41],"with":[42,186,208],"the":[43,55,60,67,136,210],"using":[45,97],"private":[46],"in-app":[47],"data.":[48],"The":[49,100,182],"difficulty":[50],"arises":[51],"due":[52],"both":[54],"resources":[58],"restricted":[61],"control":[62],"have":[65],"over":[66,174],"LLM.":[69,213],"To":[70,143],"address":[71],"this":[72,75],"issue,":[73],"in":[74],"work,":[76],"we":[77,115],"propose":[78],"LiteMoE,":[79,187],"novel":[81],"proxy":[82,98,133],"submodel":[83,108,134],"tuning":[84,155],"framework":[85],"supports":[87],"efficiently":[91],"fine-tune":[92,193],"customized":[93,194],"devices":[96],"submodels.":[99],"key":[101],"technique":[102],"behind":[103],"LiteMoE":[104,149,170],"is":[105],"post-training":[107],"extraction":[109],"method,":[110],"whereby":[111],"without":[112],"additional":[113],"retraining,":[114],"can":[116],"identify":[117],"reserve":[119],"critical":[120],"experts,":[121,126],"match":[122],"merge":[124],"moderate":[125],"extract":[128],"lightweight":[130],"effective":[132],"from":[135],"for":[139],"specific":[141],"app.":[142],"further":[144],"enhance":[145],"scalability":[146],"adaptability,":[148],"incorporates":[150],"adapter":[151],"reuse":[152],"continuous":[154],"mechanisms":[156],"handle":[158],"multi-task":[159],"requirements":[160],"evolving":[162],"user":[163],"preferences.":[164],"We":[165],"implemented":[166],"prototype":[168],"of":[169],"evaluated":[172],"it":[173],"various":[175],"MoE-based":[176],"LLMs":[177],"computing":[180],"tasks.":[181],"results":[183],"show":[184],"able":[191],"resource-limited":[197],"devices,":[198],"achieving":[199],"12.7%":[200],"accuracy":[201],"improvement":[202],"6.6\u00d7":[204],"memory":[205],"reduction":[206],"compared":[207],"operating":[209],"original":[211]},"counts_by_year":[],"updated_date":"2026-03-05T07:30:30.508283","created_date":"2026-01-29T00:00:00"}
