{"id":"https://openalex.org/W4414856582","doi":"https://doi.org/10.1109/les.2025.3616900","title":"LPC: Efficient Lossless Parameter Compression for Deploying LLM Inference on Edge Systems","display_name":"LPC: Efficient Lossless Parameter Compression for Deploying LLM Inference on Edge Systems","publication_year":2025,"publication_date":"2025-10-06","ids":{"openalex":"https://openalex.org/W4414856582","doi":"https://doi.org/10.1109/les.2025.3616900"},"language":"en","primary_location":{"id":"doi:10.1109/les.2025.3616900","is_oa":false,"landing_page_url":"https://doi.org/10.1109/les.2025.3616900","pdf_url":null,"source":{"id":"https://openalex.org/S22443479","display_name":"IEEE Embedded Systems Letters","issn_l":"1943-0663","issn":["1943-0663","1943-0671"],"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 Embedded Systems Letters","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/A5030337371","display_name":"Nachuan Wang","orcid":"https://orcid.org/0009-0000-1431-0067"},"institutions":[{"id":"https://openalex.org/I157725225","display_name":"University of Illinois Urbana-Champaign","ror":"https://ror.org/047426m28","country_code":"US","type":"education","lineage":["https://openalex.org/I157725225"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Nachuan Wang","raw_affiliation_strings":["Electrical and Computer Engineering Department, University of Illinois at Urbana–Champaign, Urbana, IL, USA","University of Illinois at Urbana-Champaign, Urbana, IL, USA"],"raw_orcid":"https://orcid.org/0009-0000-1431-0067","affiliations":[{"raw_affiliation_string":"Electrical and Computer Engineering Department, University of Illinois at Urbana–Champaign, Urbana, IL, USA","institution_ids":["https://openalex.org/I157725225"]},{"raw_affiliation_string":"University of Illinois at Urbana-Champaign, Urbana, IL, USA","institution_ids":["https://openalex.org/I157725225"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5050020235","display_name":"Hyungyo Kim","orcid":"https://orcid.org/0000-0002-1157-0127"},"institutions":[{"id":"https://openalex.org/I157725225","display_name":"University of Illinois Urbana-Champaign","ror":"https://ror.org/047426m28","country_code":"US","type":"education","lineage":["https://openalex.org/I157725225"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Hyungyo Kim","raw_affiliation_strings":["Electrical and Computer Engineering Department, University of Illinois at Urbana–Champaign, Urbana, IL, USA","University of Illinois at Urbana-Champaign, Urbana, IL, USA"],"raw_orcid":"https://orcid.org/0000-0002-1157-0127","affiliations":[{"raw_affiliation_string":"Electrical and Computer Engineering Department, University of Illinois at Urbana–Champaign, Urbana, IL, USA","institution_ids":["https://openalex.org/I157725225"]},{"raw_affiliation_string":"University of Illinois at Urbana-Champaign, Urbana, IL, USA","institution_ids":["https://openalex.org/I157725225"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5037648751","display_name":"Nam Sung Kim","orcid":"https://orcid.org/0000-0002-0442-5634"},"institutions":[{"id":"https://openalex.org/I157725225","display_name":"University of Illinois Urbana-Champaign","ror":"https://ror.org/047426m28","country_code":"US","type":"education","lineage":["https://openalex.org/I157725225"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Nam Sung Kim","raw_affiliation_strings":["Electrical and Computer Engineering Department, University of Illinois at Urbana–Champaign, Urbana, IL, USA","University of Illinois at Urbana-Champaign, Urbana, IL, USA"],"raw_orcid":"https://orcid.org/0000-0002-0442-5634","affiliations":[{"raw_affiliation_string":"Electrical and Computer Engineering Department, University of Illinois at Urbana–Champaign, Urbana, IL, USA","institution_ids":["https://openalex.org/I157725225"]},{"raw_affiliation_string":"University of Illinois at Urbana-Champaign, Urbana, IL, USA","institution_ids":["https://openalex.org/I157725225"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I157725225"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.23894453,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":"18","issue":"3","first_page":"190","last_page":"193"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10901","display_name":"Advanced Data Compression Techniques","score":0.9480999708175659,"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"}},"topics":[{"id":"https://openalex.org/T10901","display_name":"Advanced Data Compression Techniques","score":0.9480999708175659,"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/T11269","display_name":"Algorithms and Data Compression","score":0.9429000020027161,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/lossless-compression","display_name":"Lossless compression","score":0.8982999920845032},{"id":"https://openalex.org/keywords/lossy-compression","display_name":"Lossy compression","score":0.7906000018119812},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.5550000071525574},{"id":"https://openalex.org/keywords/data-compression","display_name":"Data compression","score":0.5339000225067139},{"id":"https://openalex.org/keywords/data-compression-ratio","display_name":"Data compression ratio","score":0.49149999022483826},{"id":"https://openalex.org/keywords/bandwidth","display_name":"Bandwidth (computing)","score":0.451200008392334},{"id":"https://openalex.org/keywords/compression-ratio","display_name":"Compression ratio","score":0.44850000739097595},{"id":"https://openalex.org/keywords/latency","display_name":"Latency (audio)","score":0.43549999594688416}],"concepts":[{"id":"https://openalex.org/C81081738","wikidata":"https://www.wikidata.org/wiki/Q55542","display_name":"Lossless compression","level":3,"score":0.8982999920845032},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8745999932289124},{"id":"https://openalex.org/C165021410","wikidata":"https://www.wikidata.org/wiki/Q55564","display_name":"Lossy compression","level":2,"score":0.7906000018119812},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.5550000071525574},{"id":"https://openalex.org/C78548338","wikidata":"https://www.wikidata.org/wiki/Q2493","display_name":"Data compression","level":2,"score":0.5339000225067139},{"id":"https://openalex.org/C94835093","wikidata":"https://www.wikidata.org/wiki/Q3113333","display_name":"Data compression ratio","level":5,"score":0.49149999022483826},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.45350000262260437},{"id":"https://openalex.org/C2776257435","wikidata":"https://www.wikidata.org/wiki/Q1576430","display_name":"Bandwidth (computing)","level":2,"score":0.451200008392334},{"id":"https://openalex.org/C25797200","wikidata":"https://www.wikidata.org/wiki/Q828137","display_name":"Compression ratio","level":3,"score":0.44850000739097595},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.4397999942302704},{"id":"https://openalex.org/C82876162","wikidata":"https://www.wikidata.org/wiki/Q17096504","display_name":"Latency (audio)","level":2,"score":0.43549999594688416},{"id":"https://openalex.org/C180016635","wikidata":"https://www.wikidata.org/wiki/Q2712821","display_name":"Compression (physics)","level":2,"score":0.3928000032901764},{"id":"https://openalex.org/C13481523","wikidata":"https://www.wikidata.org/wiki/Q412438","display_name":"Image compression","level":4,"score":0.388700008392334},{"id":"https://openalex.org/C2777904410","wikidata":"https://www.wikidata.org/wiki/Q7397","display_name":"Software","level":2,"score":0.3855000138282776},{"id":"https://openalex.org/C105339364","wikidata":"https://www.wikidata.org/wiki/Q2297740","display_name":"Software deployment","level":2,"score":0.3628999888896942},{"id":"https://openalex.org/C9390403","wikidata":"https://www.wikidata.org/wiki/Q3966","display_name":"Computer hardware","level":1,"score":0.3361999988555908},{"id":"https://openalex.org/C138236772","wikidata":"https://www.wikidata.org/wiki/Q25098575","display_name":"Edge device","level":3,"score":0.3156000077724457},{"id":"https://openalex.org/C165696696","wikidata":"https://www.wikidata.org/wiki/Q11287","display_name":"Exploit","level":2,"score":0.31150001287460327},{"id":"https://openalex.org/C79403827","wikidata":"https://www.wikidata.org/wiki/Q3988","display_name":"Real-time computing","level":1,"score":0.2985999882221222},{"id":"https://openalex.org/C162307627","wikidata":"https://www.wikidata.org/wiki/Q204833","display_name":"Enhanced Data Rates for GSM Evolution","level":2,"score":0.2842999994754791},{"id":"https://openalex.org/C84462506","wikidata":"https://www.wikidata.org/wiki/Q173142","display_name":"Digital signal processing","level":2,"score":0.28119999170303345},{"id":"https://openalex.org/C2778456923","wikidata":"https://www.wikidata.org/wiki/Q5337692","display_name":"Edge computing","level":3,"score":0.2687999904155731},{"id":"https://openalex.org/C125411270","wikidata":"https://www.wikidata.org/wiki/Q18653","display_name":"Encoding (memory)","level":2,"score":0.26460000872612},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.2587999999523163},{"id":"https://openalex.org/C150178126","wikidata":"https://www.wikidata.org/wiki/Q18433212","display_name":"Dynamic range compression","level":2,"score":0.25}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/les.2025.3616900","is_oa":false,"landing_page_url":"https://doi.org/10.1109/les.2025.3616900","pdf_url":null,"source":{"id":"https://openalex.org/S22443479","display_name":"IEEE Embedded Systems Letters","issn_l":"1943-0663","issn":["1943-0663","1943-0671"],"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 Embedded Systems Letters","raw_type":"journal-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G114616936","display_name":null,"funder_award_id":"RS-2025-02214649","funder_id":"https://openalex.org/F4320328359","funder_display_name":"Ministry of Science and ICT, South Korea"}],"funders":[{"id":"https://openalex.org/F4320328359","display_name":"Ministry of Science and ICT, South Korea","ror":"https://ror.org/01wpjm123"}],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"The":[0],"inference":[1,155,202],"latency":[2],"of":[3,50,109,115,152,184,204],"large":[4],"language":[5],"models":[6,185],"(LLMs)":[7],"on":[8],"edge":[9],"systems":[10],"is":[11,77],"often":[12],"bottlenecked":[13],"by":[14,30,79],"the":[15,31,48,60,80,105,121,150],"limited":[16],"memory":[17,34],"bandwidth":[18,142],"between":[19],"host":[20],"and":[21,53,62,88,120,171,188,197,208],"accelerator,":[22],"primarily":[23],"due":[24],"to":[25,38,154,175,201,206],"repeated":[26],"parameter":[27,192],"transfers":[28],"necessitated":[29],"accelerator\u2019s":[32],"restricted":[33],"capacity.":[35],"In":[36],"contrast":[37],"lossy":[39],"model":[40,51],"compression":[41,58,84,124,129],"methods":[42],"that":[43,103],"address":[44],"this":[45],"issue":[46],"at":[47,75,143],"expense":[49],"accuracy":[52,61],"time-consuming,":[54],"model-specific":[55],"tuning,":[56],"lossless":[57],"retains":[59],"offers":[63],"plug-andplay":[64],"usability,":[65],"making":[66],"it":[67],"an":[68,164],"appealing":[69],"alternative.":[70],"However,":[71],"its":[72,169],"practical":[73],"deployment":[74],"runtime":[76],"hindered":[78],"inherent":[81],"tradeoffs":[82],"among":[83],"ratio,":[85],"decompression":[86,141],"throughput,":[87],"hardware":[89,170],"overhead.":[90],"To":[91],"tackle":[92],"these":[93],"challenges,":[94],"we":[95],"present":[96],"a":[97,113,158,181],"Lossless":[98],"Parameter":[99],"Compression":[100],"(LPC)":[101],"technique":[102],"exploits":[104],"intrinsic":[106],"numerical":[107],"characteristics":[108],"LLM":[110],"parameters.":[111],"Through":[112],"combination":[114],"lightweight":[116],"value":[117],"remapping":[118],"scheme":[119],"hardware-friendly":[122],"LZ4":[123],"algorithm,":[125],"LPC":[126,153],"achieves":[127],"same":[128,140],"ratio":[130],"as":[131],"more":[132],"sophisticated":[133],"entropycoding":[134],"algorithms":[135],"(e.g.,":[136],"Deflate),":[137],"while":[138],"offering":[139],"4.5\u00d7":[144],"lower":[145],"area":[146],"cost.":[147],"We":[148],"validated":[149],"benefit":[151],"performance":[156],"in":[157,191],"real":[159],"host-accelerator":[160],"system":[161],"environment,":[162],"employing":[163],"open-source":[165],"CNN":[166],"accelerator":[167],"with":[168],"software":[172],"stack":[173],"redesigned":[174],"accommodate":[176],"PyTorch-based":[177],"LLMs.":[178],"Evaluation":[179],"across":[180],"broad":[182],"spectrum":[183],"demonstrated":[186],"29.1\u201329.4%":[187],"19.4\u201343.4%":[189],"reductions":[190],"offloading":[193],"traffic":[194],"for":[195],"BF16":[196],"INT8":[198],"formats,":[199],"leading":[200],"speedups":[203],"up":[205],"1.42\u00d7":[207],"1.76\u00d7,":[209],"respectively.":[210]},"counts_by_year":[],"updated_date":"2026-06-13T06:13:01.061226","created_date":"2025-10-10T00:00:00"}