{"id":"https://openalex.org/W4411446528","doi":"https://doi.org/10.1109/jstsp.2025.3581478","title":"Communication-Efficient Distributed On-Device LLM Inference Over Wireless Networks","display_name":"Communication-Efficient Distributed On-Device LLM Inference Over Wireless Networks","publication_year":2025,"publication_date":"2025-06-19","ids":{"openalex":"https://openalex.org/W4411446528","doi":"https://doi.org/10.1109/jstsp.2025.3581478"},"language":"en","primary_location":{"id":"doi:10.1109/jstsp.2025.3581478","is_oa":false,"landing_page_url":"https://doi.org/10.1109/jstsp.2025.3581478","pdf_url":null,"source":{"id":"https://openalex.org/S42167783","display_name":"IEEE Journal of Selected Topics in Signal Processing","issn_l":"1932-4553","issn":["1932-4553","1941-0484"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319808","host_organization_name":"Institute of Electrical and Electronics Engineers","host_organization_lineage":["https://openalex.org/P4310319808"],"host_organization_lineage_names":["Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Journal of Selected Topics in Signal Processing","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":null,"display_name":"Kai Zhang","orcid":"https://orcid.org/0009-0004-7876-5764"},"institutions":[{"id":"https://openalex.org/I200769079","display_name":"Hong Kong University of Science and Technology","ror":"https://ror.org/00q4vv597","country_code":"HK","type":"education","lineage":["https://openalex.org/I200769079"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Kai Zhang","raw_affiliation_strings":["Department of Electronic and Computer Engineering, The Hong Kong University of Science and Technology (HKUST), Kowloon, Hong Kong","Department of Electronic and Computer Engineering, The Hong Kong University of Science and Technology, Clear Water Bay, Hong Kong"],"raw_orcid":"https://orcid.org/0009-0004-7876-5764","affiliations":[{"raw_affiliation_string":"Department of Electronic and Computer Engineering, The Hong Kong University of Science and Technology (HKUST), Kowloon, Hong Kong","institution_ids":["https://openalex.org/I200769079"]},{"raw_affiliation_string":"Department of Electronic and Computer Engineering, The Hong Kong University of Science and Technology, Clear Water Bay, Hong Kong","institution_ids":["https://openalex.org/I200769079"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5044818286","display_name":"Hengtao He","orcid":"https://orcid.org/0000-0002-4659-6941"},"institutions":[{"id":"https://openalex.org/I76569877","display_name":"Southeast University","ror":"https://ror.org/04ct4d772","country_code":"CN","type":"education","lineage":["https://openalex.org/I76569877"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Hengtao He","raw_affiliation_strings":["National Mobile Communications Research Laboratory, Southeast University, Nanjing, China"],"raw_orcid":"https://orcid.org/0000-0002-4659-6941","affiliations":[{"raw_affiliation_string":"National Mobile Communications Research Laboratory, Southeast University, Nanjing, China","institution_ids":["https://openalex.org/I76569877"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5025467937","display_name":"Shenghui Song","orcid":"https://orcid.org/0000-0001-6316-8415"},"institutions":[{"id":"https://openalex.org/I200769079","display_name":"Hong Kong University of Science and Technology","ror":"https://ror.org/00q4vv597","country_code":"HK","type":"education","lineage":["https://openalex.org/I200769079"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Shenghui Song","raw_affiliation_strings":["Department of Electronic and Computer Engineering, The Hong Kong University of Science and Technology (HKUST), Kowloon, Hong Kong","Department of Electronic and Computer Engineering, The Hong Kong University of Science and Technology, Clear Water Bay, Hong Kong"],"raw_orcid":"https://orcid.org/0000-0001-6316-8415","affiliations":[{"raw_affiliation_string":"Department of Electronic and Computer Engineering, The Hong Kong University of Science and Technology (HKUST), Kowloon, Hong Kong","institution_ids":["https://openalex.org/I200769079"]},{"raw_affiliation_string":"Department of Electronic and Computer Engineering, The Hong Kong University of Science and Technology, Clear Water Bay, Hong Kong","institution_ids":["https://openalex.org/I200769079"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Jun Zhang","orcid":"https://orcid.org/0000-0002-5222-1898"},"institutions":[{"id":"https://openalex.org/I200769079","display_name":"Hong Kong University of Science and Technology","ror":"https://ror.org/00q4vv597","country_code":"HK","type":"education","lineage":["https://openalex.org/I200769079"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Jun Zhang","raw_affiliation_strings":["Department of Electronic and Computer Engineering, The Hong Kong University of Science and Technology (HKUST), Kowloon, Hong Kong","Department of Electronic and Computer Engineering, The Hong Kong University of Science and Technology, Clear Water Bay, Hong Kong"],"raw_orcid":"https://orcid.org/0000-0002-5222-1898","affiliations":[{"raw_affiliation_string":"Department of Electronic and Computer Engineering, The Hong Kong University of Science and Technology (HKUST), Kowloon, Hong Kong","institution_ids":["https://openalex.org/I200769079"]},{"raw_affiliation_string":"Department of Electronic and Computer Engineering, The Hong Kong University of Science and Technology, Clear Water Bay, Hong Kong","institution_ids":["https://openalex.org/I200769079"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5079052203","display_name":"Khaled B. Letaief","orcid":"https://orcid.org/0000-0003-2519-6401"},"institutions":[{"id":"https://openalex.org/I200769079","display_name":"Hong Kong University of Science and Technology","ror":"https://ror.org/00q4vv597","country_code":"HK","type":"education","lineage":["https://openalex.org/I200769079"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Khaled B. Letaief","raw_affiliation_strings":["Department of Electronic and Computer Engineering, The Hong Kong University of Science and Technology (HKUST), Kowloon, Hong Kong","Department of Electronic and Computer Engineering, The Hong Kong University of Science and Technology, Clear Water Bay, Hong Kong"],"raw_orcid":"https://orcid.org/0000-0003-2519-6401","affiliations":[{"raw_affiliation_string":"Department of Electronic and Computer Engineering, The Hong Kong University of Science and Technology (HKUST), Kowloon, Hong Kong","institution_ids":["https://openalex.org/I200769079"]},{"raw_affiliation_string":"Department of Electronic and Computer Engineering, The Hong Kong University of Science and Technology, Clear Water Bay, Hong Kong","institution_ids":["https://openalex.org/I200769079"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":5,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":4.3801,"has_fulltext":false,"cited_by_count":4,"citation_normalized_percentile":{"value":0.94582285,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":95,"max":98},"biblio":{"volume":"19","issue":"7","first_page":"1301","last_page":"1317"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10080","display_name":"Energy Efficient Wireless Sensor Networks","score":0.995199978351593,"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/T10080","display_name":"Energy Efficient Wireless Sensor Networks","score":0.995199978351593,"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/T11932","display_name":"Wireless Body Area Networks","score":0.993399977684021,"subfield":{"id":"https://openalex.org/subfields/2204","display_name":"Biomedical 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/T11158","display_name":"Wireless Networks and Protocols","score":0.9846000075340271,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.736819863319397},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.668304443359375},{"id":"https://openalex.org/keywords/wireless","display_name":"Wireless","score":0.5998995304107666},{"id":"https://openalex.org/keywords/wireless-network","display_name":"Wireless network","score":0.513766884803772},{"id":"https://openalex.org/keywords/computer-network","display_name":"Computer network","score":0.4869767129421234},{"id":"https://openalex.org/keywords/distributed-computing","display_name":"Distributed computing","score":0.35694101452827454},{"id":"https://openalex.org/keywords/telecommunications","display_name":"Telecommunications","score":0.2637990117073059},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.2558310627937317}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.736819863319397},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.668304443359375},{"id":"https://openalex.org/C555944384","wikidata":"https://www.wikidata.org/wiki/Q249","display_name":"Wireless","level":2,"score":0.5998995304107666},{"id":"https://openalex.org/C108037233","wikidata":"https://www.wikidata.org/wiki/Q11375","display_name":"Wireless network","level":3,"score":0.513766884803772},{"id":"https://openalex.org/C31258907","wikidata":"https://www.wikidata.org/wiki/Q1301371","display_name":"Computer network","level":1,"score":0.4869767129421234},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.35694101452827454},{"id":"https://openalex.org/C76155785","wikidata":"https://www.wikidata.org/wiki/Q418","display_name":"Telecommunications","level":1,"score":0.2637990117073059},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.2558310627937317}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1109/jstsp.2025.3581478","is_oa":false,"landing_page_url":"https://doi.org/10.1109/jstsp.2025.3581478","pdf_url":null,"source":{"id":"https://openalex.org/S42167783","display_name":"IEEE Journal of Selected Topics in Signal Processing","issn_l":"1932-4553","issn":["1932-4553","1941-0484"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319808","host_organization_name":"Institute of Electrical and Electronics Engineers","host_organization_lineage":["https://openalex.org/P4310319808"],"host_organization_lineage_names":["Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Journal of Selected Topics in Signal Processing","raw_type":"journal-article"},{"id":"pmh:oai:repository.hkust.edu.hk:1783.1-157553","is_oa":false,"landing_page_url":"http://repository.hkust.edu.hk/ir/Record/1783.1-157553","pdf_url":null,"source":{"id":"https://openalex.org/S4306401796","display_name":"Rare & Special e-Zone (The Hong Kong University of Science and Technology)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I200769079","host_organization_name":"Hong Kong University of Science and Technology","host_organization_lineage":["https://openalex.org/I200769079"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Article"}],"best_oa_location":null,"sustainable_development_goals":[{"score":0.44999998807907104,"display_name":"Affordable and clean energy","id":"https://metadata.un.org/sdg/7"}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W2055243143","https://openalex.org/W1986418932","https://openalex.org/W2357796999","https://openalex.org/W4321636575","https://openalex.org/W2111375262","https://openalex.org/W2741131631","https://openalex.org/W2045526782","https://openalex.org/W2317728013","https://openalex.org/W2156919374","https://openalex.org/W4252076617"],"abstract_inverted_index":{"Large":[0],"language":[1],"models":[2],"(LLMs)":[3],"have":[4],"demonstrated":[5],"remarkable":[6],"success":[7],"across":[8,57,80],"various":[9],"application":[10],"domains,":[11],"but":[12],"their":[13],"enormous":[14],"sizes":[15],"and":[16,123,133,152,200],"computational":[17,118],"demands":[18],"pose":[19],"significant":[20,85],"challenges":[21],"for":[22,61,75],"deployment":[23],"on":[24],"resource-constrained":[25],"edge":[26,59,121],"devices.":[27],"To":[28,88,114],"address":[29],"this":[30,90],"issue,":[31],"we":[32,92,127,153,163],"propose":[33,93,154],"a":[34,129,173],"novel":[35],"distributed":[36],"on-device":[37],"LLM":[38,56],"inference":[39,197,202],"framework":[40,188],"that":[41,99,165,185],"leverages":[42],"tensor":[43,68],"parallelism":[44,69],"to":[45,109,137,159,172,195],"partition":[46],"the":[47,71,101,116,139,166,177,186],"neural":[48],"network":[49],"tensors":[50],"(e.g.,":[51],"weight":[52],"matrices)":[53],"of":[54,105,120,176],"one":[55],"multiple":[58],"devices":[60,122],"collaborative":[62],"inference.":[63],"A":[64],"key":[65],"challenge":[66],"in":[67],"is":[70,150],"frequent":[72],"all-reduce":[73,112],"operations":[74],"aggregating":[76],"intermediate":[77],"layer":[78],"outputs":[79],"participating":[81],"devices,":[82],"which":[83],"incurs":[84],"communication":[86,125],"overhead.":[87],"alleviate":[89],"bottleneck,":[91],"an":[94,155],"over-the-air":[95],"computation":[96],"(AirComp)":[97],"approach":[98],"harnesses":[100],"analog":[102],"superposition":[103],"property":[104],"wireless":[106],"multiple-access":[107],"channels":[108],"perform":[110],"fast":[111],"steps.":[113],"utilize":[115],"heterogeneous":[117],"capabilities":[119],"mitigate":[124],"distortions,":[126],"investigate":[128],"joint":[130],"model":[131],"assignment":[132],"transceiver":[134],"optimization":[135,148],"problem":[136,149],"minimize":[138],"average":[140],"transmission":[141],"error.":[142],"The":[143],"resulting":[144],"mixed-timescale":[145],"stochastic":[146],"non-convex":[147],"intractable,":[151],"efficient":[156],"two-stage":[157],"algorithm":[158,168],"solve":[160],"it.":[161],"Moreover,":[162],"prove":[164],"proposed":[167,187],"converges":[169],"almost":[170],"surely":[171],"stationary":[174],"point":[175],"original":[178],"problem.":[179],"Comprehensive":[180],"simulation":[181],"results":[182],"will":[183],"show":[184],"outperforms":[189],"existing":[190],"benchmark":[191],"schemes,":[192],"achieving":[193],"up":[194],"5x":[196],"speed":[198],"acceleration":[199],"improving":[201],"accuracy.":[203]},"counts_by_year":[{"year":2026,"cited_by_count":1},{"year":2025,"cited_by_count":3}],"updated_date":"2026-06-13T07:54:00.901334","created_date":"2025-10-10T00:00:00"}