{"id":"https://openalex.org/W7084049098","doi":"https://doi.org/10.1109/infocom55648.2025.11044514","title":"FLM-TopK: Expediting Federated Large Language Model Tuning by Sparsifying Intervalized Gradients","display_name":"FLM-TopK: Expediting Federated Large Language Model Tuning by Sparsifying Intervalized Gradients","publication_year":2025,"publication_date":"2025-05-19","ids":{"openalex":"https://openalex.org/W7084049098","doi":"https://doi.org/10.1109/infocom55648.2025.11044514"},"language":"en","primary_location":{"id":"doi:10.1109/infocom55648.2025.11044514","is_oa":false,"landing_page_url":"https://doi.org/10.1109/infocom55648.2025.11044514","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE INFOCOM 2025 - IEEE Conference on Computer Communications","raw_type":"proceedings-article"},"type":"conference-paper","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":null,"display_name":"Wenqi Qiu","orcid":null},"institutions":[{"id":"https://openalex.org/I180726961","display_name":"Shenzhen University","ror":"https://ror.org/01vy4gh70","country_code":"CN","type":"education","lineage":["https://openalex.org/I180726961"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Wenqi Qiu","raw_affiliation_strings":["College of Computer Science and Software Engineering, Shenzhen University,Shenzhen,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"College of Computer Science and Software Engineering, Shenzhen University,Shenzhen,China","institution_ids":["https://openalex.org/I180726961"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Yipeng Zhou","orcid":null},"institutions":[{"id":"https://openalex.org/I99043593","display_name":"Macquarie University","ror":"https://ror.org/01sf06y89","country_code":"AU","type":"education","lineage":["https://openalex.org/I99043593"]}],"countries":["AU"],"is_corresponding":false,"raw_author_name":"Yipeng Zhou","raw_affiliation_strings":["School of Computing, Macquarie University,Faculty of Science and Engineering,Sydney,Australia"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Computing, Macquarie University,Faculty of Science and Engineering,Sydney,Australia","institution_ids":["https://openalex.org/I99043593"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Jinzhi Wang","orcid":null},"institutions":[{"id":"https://openalex.org/I4210136793","display_name":"Peng Cheng Laboratory","ror":"https://ror.org/03qdqbt06","country_code":"CN","type":"facility","lineage":["https://openalex.org/I4210136793"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Jinzhi Wang","raw_affiliation_strings":["Guangdong Laboratory of Artificial Intelligence and Digital Economy (SZ),Shenzhen,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Guangdong Laboratory of Artificial Intelligence and Digital Economy (SZ),Shenzhen,China","institution_ids":["https://openalex.org/I4210136793"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Quan Z. Sheng","orcid":null},"institutions":[{"id":"https://openalex.org/I99043593","display_name":"Macquarie University","ror":"https://ror.org/01sf06y89","country_code":"AU","type":"education","lineage":["https://openalex.org/I99043593"]}],"countries":["AU"],"is_corresponding":false,"raw_author_name":"Quan Z. Sheng","raw_affiliation_strings":["School of Computing, Macquarie University,Faculty of Science and Engineering,Sydney,Australia"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Computing, Macquarie University,Faculty of Science and Engineering,Sydney,Australia","institution_ids":["https://openalex.org/I99043593"]}]},{"author_position":"last","author":{"id":null,"display_name":"Laizhong Cui","orcid":null},"institutions":[{"id":"https://openalex.org/I180726961","display_name":"Shenzhen University","ror":"https://ror.org/01vy4gh70","country_code":"CN","type":"education","lineage":["https://openalex.org/I180726961"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Laizhong Cui","raw_affiliation_strings":["College of Computer Science and Software Engineering, Shenzhen University,Shenzhen,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"College of Computer Science and Software Engineering, Shenzhen University,Shenzhen,China","institution_ids":["https://openalex.org/I180726961"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":3,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":3,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"10"},"is_retracted":false,"is_paratext":false,"is_xpac":true,"primary_topic":{"id":"https://openalex.org/T10091","display_name":"Malaria Research and Control","score":0.9828000068664551,"subfield":{"id":"https://openalex.org/subfields/2739","display_name":"Public Health, Environmental and Occupational Health"},"field":{"id":"https://openalex.org/fields/27","display_name":"Medicine"},"domain":{"id":"https://openalex.org/domains/4","display_name":"Health Sciences"}},"topics":[{"id":"https://openalex.org/T10091","display_name":"Malaria Research and Control","score":0.9828000068664551,"subfield":{"id":"https://openalex.org/subfields/2739","display_name":"Public Health, Environmental and Occupational Health"},"field":{"id":"https://openalex.org/fields/27","display_name":"Medicine"},"domain":{"id":"https://openalex.org/domains/4","display_name":"Health Sciences"}},{"id":"https://openalex.org/T10695","display_name":"Parasites and Host Interactions","score":0.002099999925121665,"subfield":{"id":"https://openalex.org/subfields/2405","display_name":"Parasitology"},"field":{"id":"https://openalex.org/fields/24","display_name":"Immunology and Microbiology"},"domain":{"id":"https://openalex.org/domains/1","display_name":"Life Sciences"}},{"id":"https://openalex.org/T10209","display_name":"Global Maternal and Child Health","score":0.0012000000569969416,"subfield":{"id":"https://openalex.org/subfields/2735","display_name":"Pediatrics, Perinatology and Child Health"},"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/expediting","display_name":"Expediting","score":0.7840999960899353},{"id":"https://openalex.org/keywords/quantization","display_name":"Quantization (signal processing)","score":0.6212000250816345},{"id":"https://openalex.org/keywords/language-model","display_name":"Language model","score":0.6148999929428101},{"id":"https://openalex.org/keywords/overhead","display_name":"Overhead (engineering)","score":0.6072999835014343},{"id":"https://openalex.org/keywords/gaussian","display_name":"Gaussian","score":0.47110000252723694},{"id":"https://openalex.org/keywords/interval","display_name":"Interval (graph theory)","score":0.46059998869895935},{"id":"https://openalex.org/keywords/gaussian-network-model","display_name":"Gaussian network model","score":0.41200000047683716},{"id":"https://openalex.org/keywords/reduction","display_name":"Reduction (mathematics)","score":0.3894999921321869}],"concepts":[{"id":"https://openalex.org/C134448949","wikidata":"https://www.wikidata.org/wiki/Q1384274","display_name":"Expediting","level":2,"score":0.7840999960899353},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.771399974822998},{"id":"https://openalex.org/C28855332","wikidata":"https://www.wikidata.org/wiki/Q198099","display_name":"Quantization (signal processing)","level":2,"score":0.6212000250816345},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.6148999929428101},{"id":"https://openalex.org/C2779960059","wikidata":"https://www.wikidata.org/wiki/Q7113681","display_name":"Overhead (engineering)","level":2,"score":0.6072999835014343},{"id":"https://openalex.org/C163716315","wikidata":"https://www.wikidata.org/wiki/Q901177","display_name":"Gaussian","level":2,"score":0.47110000252723694},{"id":"https://openalex.org/C2778067643","wikidata":"https://www.wikidata.org/wiki/Q166507","display_name":"Interval (graph theory)","level":2,"score":0.46059998869895935},{"id":"https://openalex.org/C166550679","wikidata":"https://www.wikidata.org/wiki/Q263400","display_name":"Gaussian network model","level":3,"score":0.41200000047683716},{"id":"https://openalex.org/C111335779","wikidata":"https://www.wikidata.org/wiki/Q3454686","display_name":"Reduction (mathematics)","level":2,"score":0.3894999921321869},{"id":"https://openalex.org/C77618280","wikidata":"https://www.wikidata.org/wiki/Q1155772","display_name":"Scheme (mathematics)","level":2,"score":0.35530000925064087},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.34709998965263367},{"id":"https://openalex.org/C61326573","wikidata":"https://www.wikidata.org/wiki/Q1496376","display_name":"Gaussian process","level":3,"score":0.33739998936653137},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.3353999853134155},{"id":"https://openalex.org/C137836250","wikidata":"https://www.wikidata.org/wiki/Q984063","display_name":"Optimization problem","level":2,"score":0.3352999985218048},{"id":"https://openalex.org/C139807058","wikidata":"https://www.wikidata.org/wiki/Q352374","display_name":"Adaptation (eye)","level":2,"score":0.3352000117301941},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.3255000114440918},{"id":"https://openalex.org/C198082294","wikidata":"https://www.wikidata.org/wiki/Q3399648","display_name":"Position (finance)","level":2,"score":0.31619998812675476},{"id":"https://openalex.org/C79403827","wikidata":"https://www.wikidata.org/wiki/Q3988","display_name":"Real-time computing","level":1,"score":0.3122999966144562},{"id":"https://openalex.org/C165064840","wikidata":"https://www.wikidata.org/wiki/Q1321061","display_name":"Matching (statistics)","level":2,"score":0.2962000072002411},{"id":"https://openalex.org/C153258448","wikidata":"https://www.wikidata.org/wiki/Q1199743","display_name":"Gradient descent","level":3,"score":0.28700000047683716},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.2761000096797943},{"id":"https://openalex.org/C86610423","wikidata":"https://www.wikidata.org/wiki/Q1925081","display_name":"Metamodeling","level":2,"score":0.26840001344680786},{"id":"https://openalex.org/C25797200","wikidata":"https://www.wikidata.org/wiki/Q828137","display_name":"Compression ratio","level":3,"score":0.25839999318122864},{"id":"https://openalex.org/C2780138299","wikidata":"https://www.wikidata.org/wiki/Q3404265","display_name":"Privilege (computing)","level":2,"score":0.25429999828338623},{"id":"https://openalex.org/C192209626","wikidata":"https://www.wikidata.org/wiki/Q190909","display_name":"Focus (optics)","level":2,"score":0.25290000438690186},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.25200000405311584}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/infocom55648.2025.11044514","is_oa":false,"landing_page_url":"https://doi.org/10.1109/infocom55648.2025.11044514","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE INFOCOM 2025 - IEEE Conference on Computer Communications","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G4167228550","display_name":null,"funder_award_id":"U23B2026,62372305","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":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"The":[0,214],"past":[1],"few":[2],"years":[3],"have":[4],"witnessed":[5],"the":[6,47,51,62,65,112,138,151,157,162,172,178,182,190,202],"unprecedented":[7],"capability":[8],"of":[9,64,94,153,204],"large":[10],"language":[11],"models":[12],"(LLMs).":[13],"To":[14,68,144,200],"adapt":[15],"LLMs":[16,40],"with":[17,41,105,116],"various":[18],"downstream":[19],"tasks,":[20],"fine-tuning":[21],"methods,":[22],"e.g.,":[23],"Low-Rank":[24],"Adaptation":[25],"(LoRA),":[26],"are":[27],"proposed":[28],"to":[29,100,170],"efficiently":[30],"tune":[31],"LLMs.":[32],"Meanwhile,":[33],"federated":[34,48,102],"LLM":[35,103,114],"tuning":[36,49,104],"emerges":[37],"for":[38,78,89,131],"refining":[39],"clients":[42,54],"owning":[43],"private":[44],"data.":[45],"In":[46,97],"process,":[50],"server":[52],"and":[53,181,228],"frequently":[55],"exchange":[56],"fine-tune":[57],"gradients":[58,80,123,132],"via":[59],"Internet,":[60],"giving":[61],"rise":[63],"communication":[66,232],"challenge.":[67],"overcome":[69],"this":[70,98],"challenge,":[71],"most":[72],"existing":[73],"works":[74],"employ":[75],"quantization":[76],"methods":[77,83],"compressing":[79],"because":[81],"sparsification":[82,183],"like":[84],"TopK":[85,117,127],"incur":[86],"heavy":[87],"overhead":[88,139],"transmitting":[90],"position":[91],"IDs":[92],"(PIDs)":[93],"sparsified":[95],"gradients.":[96],"work,":[99],"expedite":[101],"a":[106],"higher":[107],"compression":[108,173],"rate,":[109],"we":[110,148,165,206],"design":[111],"Federated":[113],"Tuning":[115],"(FLM-TopK)":[118],"algorithm.":[119],"Specifically,":[120],"FLM-TopK":[121,218],"intervalizes":[122],"before":[124],"compression.":[125],"Then,":[126],"is":[128,142],"separately":[129],"applied":[130],"in":[133,226,231],"each":[134],"interval":[135,179],"so":[136],"that":[137,189,217],"representing":[140],"PIDs":[141],"constrained.":[143],"optimize":[145],"our":[146],"algorithm,":[147],"empirically":[149],"study":[150],"distribution":[152],"gradients,":[154],"which":[155],"obeys":[156],"Gaussian":[158,163],"distribution.":[159],"Based":[160],"on":[161,210],"distribution,":[164],"establish":[166],"an":[167],"optimization":[168],"problem":[169,192],"minimize":[171],"error":[174],"by":[175,197],"jointly":[176],"optimizing":[177],"size":[180],"rate":[184],"per":[185],"interval.":[186],"We":[187],"prove":[188],"non-convex":[191],"can":[193],"be":[194],"approximately":[195],"solved":[196],"alternating":[198],"optimization.":[199],"demonstrate":[201,216],"superiority":[203],"FLM-TopK,":[205],"conduct":[207],"extensive":[208],"experiments":[209],"nine":[211],"public":[212],"datasets.":[213],"results":[215],"significantly":[219],"outperforms":[220],"SOTA":[221],"baselines,":[222],"achieving":[223],"6.42%-18.87%":[224],"improvement":[225],"accuracy":[227],"17.07%-44.44%":[229],"reduction":[230],"traffic.":[233]},"counts_by_year":[{"year":2026,"cited_by_count":3}],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-10-10T00:00:00"}