{"id":"https://openalex.org/W4414539726","doi":"https://doi.org/10.1109/icc52391.2025.11161391","title":"Distributed on-Device LLM Inference with Over-the-Air Computation","display_name":"Distributed on-Device LLM Inference with Over-the-Air Computation","publication_year":2025,"publication_date":"2025-06-08","ids":{"openalex":"https://openalex.org/W4414539726","doi":"https://doi.org/10.1109/icc52391.2025.11161391"},"language":"en","primary_location":{"id":"doi:10.1109/icc52391.2025.11161391","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icc52391.2025.11161391","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICC 2025 - IEEE International Conference on Communications","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/A5100324056","display_name":"Kai Zhang","orcid":"https://orcid.org/0000-0003-3850-5429"},"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":true,"raw_author_name":"Kai Zhang","raw_affiliation_strings":["The Hong Kong University of Science and Technology,Department of Electronic and Computer Engineering"],"affiliations":[{"raw_affiliation_string":"The Hong Kong University of Science and Technology,Department of Electronic and Computer Engineering","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/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":"Hengtao He","raw_affiliation_strings":["The Hong Kong University of Science and Technology,Department of Electronic and Computer Engineering"],"affiliations":[{"raw_affiliation_string":"The Hong Kong University of Science and Technology,Department of Electronic and Computer Engineering","institution_ids":["https://openalex.org/I200769079"]}]},{"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":["The Hong Kong University of Science and Technology,Department of Electronic and Computer Engineering"],"affiliations":[{"raw_affiliation_string":"The Hong Kong University of Science and Technology,Department of Electronic and Computer Engineering","institution_ids":["https://openalex.org/I200769079"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100433167","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":["The Hong Kong University of Science and Technology,Department of Electronic and Computer Engineering"],"affiliations":[{"raw_affiliation_string":"The Hong Kong University of Science and Technology,Department of Electronic and Computer Engineering","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":["The Hong Kong University of Science and Technology,Department of Electronic and Computer Engineering"],"affiliations":[{"raw_affiliation_string":"The Hong Kong University of Science and Technology,Department of Electronic and Computer Engineering","institution_ids":["https://openalex.org/I200769079"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":5,"corresponding_author_ids":["https://openalex.org/A5100324056"],"corresponding_institution_ids":["https://openalex.org/I200769079"],"apc_list":null,"apc_paid":null,"fwci":3.0747,"has_fulltext":false,"cited_by_count":2,"citation_normalized_percentile":{"value":0.92432308,"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":"6291","last_page":"6296"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10522","display_name":"Medical Imaging Techniques and Applications","score":0.7621999979019165,"subfield":{"id":"https://openalex.org/subfields/2741","display_name":"Radiology, Nuclear Medicine and Imaging"},"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/T10522","display_name":"Medical Imaging Techniques and Applications","score":0.7621999979019165,"subfield":{"id":"https://openalex.org/subfields/2741","display_name":"Radiology, Nuclear Medicine and Imaging"},"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/T10036","display_name":"Advanced Neural Network Applications","score":0.727400004863739,"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/T10688","display_name":"Image and Signal Denoising Methods","score":0.7129999995231628,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.6958000063896179},{"id":"https://openalex.org/keywords/computation","display_name":"Computation","score":0.5533999800682068},{"id":"https://openalex.org/keywords/latency","display_name":"Latency (audio)","score":0.47200000286102295},{"id":"https://openalex.org/keywords/scalability","display_name":"Scalability","score":0.4666000008583069},{"id":"https://openalex.org/keywords/enhanced-data-rates-for-gsm-evolution","display_name":"Enhanced Data Rates for GSM Evolution","score":0.44749999046325684},{"id":"https://openalex.org/keywords/edge-device","display_name":"Edge device","score":0.4336000084877014},{"id":"https://openalex.org/keywords/relaxation","display_name":"Relaxation (psychology)","score":0.4106999933719635},{"id":"https://openalex.org/keywords/convex-optimization","display_name":"Convex optimization","score":0.41029998660087585},{"id":"https://openalex.org/keywords/optimization-problem","display_name":"Optimization problem","score":0.4023999869823456},{"id":"https://openalex.org/keywords/applications-of-artificial-intelligence","display_name":"Applications of artificial intelligence","score":0.383899986743927}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7551000118255615},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.6958000063896179},{"id":"https://openalex.org/C45374587","wikidata":"https://www.wikidata.org/wiki/Q12525525","display_name":"Computation","level":2,"score":0.5533999800682068},{"id":"https://openalex.org/C82876162","wikidata":"https://www.wikidata.org/wiki/Q17096504","display_name":"Latency (audio)","level":2,"score":0.47200000286102295},{"id":"https://openalex.org/C48044578","wikidata":"https://www.wikidata.org/wiki/Q727490","display_name":"Scalability","level":2,"score":0.4666000008583069},{"id":"https://openalex.org/C162307627","wikidata":"https://www.wikidata.org/wiki/Q204833","display_name":"Enhanced Data Rates for GSM Evolution","level":2,"score":0.44749999046325684},{"id":"https://openalex.org/C138236772","wikidata":"https://www.wikidata.org/wiki/Q25098575","display_name":"Edge device","level":3,"score":0.4336000084877014},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.4287000000476837},{"id":"https://openalex.org/C2776029896","wikidata":"https://www.wikidata.org/wiki/Q3935810","display_name":"Relaxation (psychology)","level":2,"score":0.4106999933719635},{"id":"https://openalex.org/C157972887","wikidata":"https://www.wikidata.org/wiki/Q463359","display_name":"Convex optimization","level":3,"score":0.41029998660087585},{"id":"https://openalex.org/C137836250","wikidata":"https://www.wikidata.org/wiki/Q984063","display_name":"Optimization problem","level":2,"score":0.4023999869823456},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.38589999079704285},{"id":"https://openalex.org/C157170001","wikidata":"https://www.wikidata.org/wiki/Q4781507","display_name":"Applications of artificial intelligence","level":2,"score":0.383899986743927},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3815000057220459},{"id":"https://openalex.org/C27753989","wikidata":"https://www.wikidata.org/wiki/Q284885","display_name":"Superposition principle","level":2,"score":0.376800000667572},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.3718999922275543},{"id":"https://openalex.org/C130120984","wikidata":"https://www.wikidata.org/wiki/Q2835898","display_name":"Distributed algorithm","level":2,"score":0.3698999881744385},{"id":"https://openalex.org/C555944384","wikidata":"https://www.wikidata.org/wiki/Q249","display_name":"Wireless","level":2,"score":0.36250001192092896},{"id":"https://openalex.org/C4679612","wikidata":"https://www.wikidata.org/wiki/Q866298","display_name":"Aggregate (composite)","level":2,"score":0.35600000619888306},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.3418000042438507},{"id":"https://openalex.org/C854659","wikidata":"https://www.wikidata.org/wiki/Q1859284","display_name":"Message passing","level":2,"score":0.32670000195503235},{"id":"https://openalex.org/C108037233","wikidata":"https://www.wikidata.org/wiki/Q11375","display_name":"Wireless network","level":3,"score":0.320499986410141},{"id":"https://openalex.org/C105339364","wikidata":"https://www.wikidata.org/wiki/Q2297740","display_name":"Software deployment","level":2,"score":0.3163999915122986},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.3018999993801117},{"id":"https://openalex.org/C761482","wikidata":"https://www.wikidata.org/wiki/Q118093","display_name":"Transmission (telecommunications)","level":2,"score":0.296099990606308},{"id":"https://openalex.org/C155281189","wikidata":"https://www.wikidata.org/wiki/Q3518150","display_name":"Tensor (intrinsic definition)","level":2,"score":0.29179999232292175},{"id":"https://openalex.org/C206688291","wikidata":"https://www.wikidata.org/wiki/Q7617819","display_name":"Stochastic gradient descent","level":3,"score":0.2897000014781952},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.2874000072479248},{"id":"https://openalex.org/C2778456923","wikidata":"https://www.wikidata.org/wiki/Q5337692","display_name":"Edge computing","level":3,"score":0.2840999960899353},{"id":"https://openalex.org/C189950617","wikidata":"https://www.wikidata.org/wiki/Q937228","display_name":"Property (philosophy)","level":2,"score":0.2833000123500824},{"id":"https://openalex.org/C147764199","wikidata":"https://www.wikidata.org/wiki/Q6865248","display_name":"Minification","level":2,"score":0.2793000042438507},{"id":"https://openalex.org/C99498987","wikidata":"https://www.wikidata.org/wiki/Q2210247","display_name":"Noise (video)","level":3,"score":0.2766000032424927},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.26510000228881836},{"id":"https://openalex.org/C24590314","wikidata":"https://www.wikidata.org/wiki/Q336038","display_name":"Wireless sensor network","level":2,"score":0.2621999979019165},{"id":"https://openalex.org/C194387892","wikidata":"https://www.wikidata.org/wiki/Q1747770","display_name":"Stochastic optimization","level":2,"score":0.2619999945163727},{"id":"https://openalex.org/C112680207","wikidata":"https://www.wikidata.org/wiki/Q714886","display_name":"Regular polygon","level":2,"score":0.258899986743927},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.2522999942302704}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1109/icc52391.2025.11161391","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icc52391.2025.11161391","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICC 2025 - IEEE International Conference on Communications","raw_type":"proceedings-article"},{"id":"pmh:oai:repository.hkust.edu.hk:1783.1-151438","is_oa":false,"landing_page_url":"http://repository.hkust.edu.hk/ir/Record/1783.1-151438","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":"Conference paper"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":12,"referenced_works":["https://openalex.org/W1996215314","https://openalex.org/W2087039524","https://openalex.org/W2784991221","https://openalex.org/W2887810578","https://openalex.org/W2969519626","https://openalex.org/W3212941463","https://openalex.org/W3214462705","https://openalex.org/W4382246105","https://openalex.org/W4390771941","https://openalex.org/W4400447774","https://openalex.org/W4405934565","https://openalex.org/W4409474237"],"related_works":[],"abstract_inverted_index":{"Large":[0],"language":[1],"models":[2],"(LLMs)":[3],"have":[4],"achieved":[5],"remarkable":[6],"success":[7],"across":[8,74],"various":[9],"artificial":[10],"intelligence":[11],"tasks.":[12],"However,":[13],"their":[14],"enormous":[15],"sizes":[16],"and":[17,121,144],"computational":[18],"demands":[19],"pose":[20],"significant":[21],"challenges":[22],"for":[23,59,173],"the":[24,96,111,156],"deployment":[25],"on":[26,42],"edge":[27,57,175],"devices.":[28,176],"To":[29,84,109],"address":[30],"this":[31,86],"issue,":[32],"we":[33,88,116,136],"present":[34],"a":[35,129,138],"distributed":[36,168],"on-device":[37],"LLM":[38,170],"inference":[39,161,171],"framework":[40],"based":[41],"tensor":[43,63],"parallelism,":[44],"which":[45,124],"partitions":[46],"neural":[47],"network":[48],"tensors":[49],"(e.g.,":[50],"weight":[51],"matrices)":[52],"of":[53,100],"LLMs":[54],"among":[55],"multiple":[56],"devices":[58,76],"collaborative":[60],"inference.":[61],"Nevertheless,":[62],"parallelism":[64],"involves":[65],"frequent":[66],"all-reduce":[67,107],"operations":[68],"to":[69,104],"aggregate":[70],"intermediate":[71],"layer":[72],"outputs":[73],"participating":[75],"during":[77],"inference,":[78],"resulting":[79],"in":[80],"substantial":[81],"communication":[82],"overhead.":[83],"mitigate":[85],"bottleneck,":[87],"propose":[89],"an":[90],"over-the-air":[91],"computation":[92],"method":[93],"that":[94,155],"leverages":[95],"analog":[97],"superposition":[98],"property":[99],"wireless":[101],"multipleaccess":[102],"channels":[103],"facilitate":[105],"fast":[106],"operations.":[108],"minimize":[110],"average":[112],"transmission":[113],"mean-squared":[114],"error,":[115],"investigate":[117],"joint":[118],"model":[119],"assignment":[120],"transceiver":[122],"optimization,":[123],"can":[125],"be":[126],"formulated":[127],"as":[128],"mixed-timescale":[130,139],"stochastic":[131,145],"non-convex":[132],"optimization":[133],"problem.":[134],"Then,":[135],"develop":[137],"algorithm":[140],"leveraging":[141],"semidefinite":[142],"relaxation":[143],"successive":[146],"convex":[147],"approximation":[148],"methods.":[149],"Comprehensive":[150],"simulation":[151],"results":[152],"will":[153],"show":[154],"proposed":[157],"approach":[158],"significantly":[159],"reduces":[160],"latency":[162],"while":[163],"improving":[164],"accuracy.":[165],"This":[166],"makes":[167],"ondevice":[169],"practical":[172],"resource-constrained":[174]},"counts_by_year":[{"year":2026,"cited_by_count":1},{"year":2025,"cited_by_count":1}],"updated_date":"2026-03-07T16:01:11.037858","created_date":"2025-10-10T00:00:00"}