{"id":"https://openalex.org/W7133551067","doi":"https://doi.org/10.1109/hpca68181.2026.11408577","title":"LiLo: Harnessing the on-Chip Accelerators in Intel CPUs for Compressed LLM Inference Acceleration","display_name":"LiLo: Harnessing the on-Chip Accelerators in Intel CPUs for Compressed LLM Inference Acceleration","publication_year":2026,"publication_date":"2026-01-31","ids":{"openalex":"https://openalex.org/W7133551067","doi":"https://doi.org/10.1109/hpca68181.2026.11408577"},"language":null,"primary_location":{"id":"doi:10.1109/hpca68181.2026.11408577","is_oa":false,"landing_page_url":"https://doi.org/10.1109/hpca68181.2026.11408577","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2026 IEEE International Symposium on High Performance Computer Architecture (HPCA)","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/A5084753791","display_name":"Hyo-yeon Kim","orcid":null},"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":true,"raw_author_name":"Hyungyo Kim","raw_affiliation_strings":["University of Illinois at Urbana-Champaign"],"affiliations":[{"raw_affiliation_string":"University of Illinois at Urbana-Champaign","institution_ids":["https://openalex.org/I157725225"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5004543720","display_name":"Qirong Xia","orcid":null},"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":"Qirong Xia","raw_affiliation_strings":["University of Illinois at Urbana-Champaign"],"affiliations":[{"raw_affiliation_string":"University of Illinois at Urbana-Champaign","institution_ids":["https://openalex.org/I157725225"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5033906953","display_name":"Jinghan Huang","orcid":"https://orcid.org/0009-0009-2314-2734"},"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":"Jinghan Huang","raw_affiliation_strings":["University of Illinois at Urbana-Champaign"],"affiliations":[{"raw_affiliation_string":"University of Illinois at Urbana-Champaign","institution_ids":["https://openalex.org/I157725225"]}]},{"author_position":"middle","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":["University of Illinois at Urbana-Champaign"],"affiliations":[{"raw_affiliation_string":"University of Illinois at Urbana-Champaign","institution_ids":["https://openalex.org/I157725225"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5041436589","display_name":"Younjoo Lee","orcid":"https://orcid.org/0000-0002-2869-9680"},"institutions":[{"id":"https://openalex.org/I139264467","display_name":"Seoul National University","ror":"https://ror.org/04h9pn542","country_code":"KR","type":"education","lineage":["https://openalex.org/I139264467"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Younjoo Lee","raw_affiliation_strings":["Seoul National University"],"affiliations":[{"raw_affiliation_string":"Seoul National University","institution_ids":["https://openalex.org/I139264467"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5125243766","display_name":"J Ahn","orcid":null},"institutions":[{"id":"https://openalex.org/I139264467","display_name":"Seoul National University","ror":"https://ror.org/04h9pn542","country_code":"KR","type":"education","lineage":["https://openalex.org/I139264467"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Jung Ho Ahn","raw_affiliation_strings":["Seoul National University"],"affiliations":[{"raw_affiliation_string":"Seoul National University","institution_ids":["https://openalex.org/I139264467"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5024446685","display_name":"Wajdi K. Feghali","orcid":null},"institutions":[{"id":"https://openalex.org/I4210158342","display_name":"Intel (United Kingdom)","ror":"https://ror.org/058cxws58","country_code":"GB","type":"company","lineage":["https://openalex.org/I1343180700","https://openalex.org/I4210158342"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Wajdi K Feghali","raw_affiliation_strings":["Intel Corporation"],"affiliations":[{"raw_affiliation_string":"Intel Corporation","institution_ids":["https://openalex.org/I4210158342"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5036203329","display_name":"R. Wang","orcid":null},"institutions":[{"id":"https://openalex.org/I4210158342","display_name":"Intel (United Kingdom)","ror":"https://ror.org/058cxws58","country_code":"GB","type":"company","lineage":["https://openalex.org/I1343180700","https://openalex.org/I4210158342"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Ren Wang","raw_affiliation_strings":["Intel Corporation"],"affiliations":[{"raw_affiliation_string":"Intel Corporation","institution_ids":["https://openalex.org/I4210158342"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5068829361","display_name":"Namsung Kim","orcid":null},"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":["University of Illinois at Urbana-Champaign"],"affiliations":[{"raw_affiliation_string":"University of Illinois at Urbana-Champaign","institution_ids":["https://openalex.org/I157725225"]}]}],"institutions":[],"countries_distinct_count":3,"institutions_distinct_count":9,"corresponding_author_ids":["https://openalex.org/A5084753791"],"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":0.92667099,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"17"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.2558000087738037,"subfield":{"id":"https://openalex.org/subfields/1708","display_name":"Hardware and Architecture"},"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/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.2558000087738037,"subfield":{"id":"https://openalex.org/subfields/1708","display_name":"Hardware and Architecture"},"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/T10363","display_name":"Low-power high-performance VLSI design","score":0.10890000313520432,"subfield":{"id":"https://openalex.org/subfields/2208","display_name":"Electrical and Electronic 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/T10904","display_name":"Embedded Systems Design Techniques","score":0.0982000008225441,"subfield":{"id":"https://openalex.org/subfields/1708","display_name":"Hardware and Architecture"},"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/acceleration","display_name":"Acceleration","score":0.5961999893188477},{"id":"https://openalex.org/keywords/key","display_name":"Key (lock)","score":0.2696000039577484},{"id":"https://openalex.org/keywords/interface","display_name":"Interface (matter)","score":0.2669000029563904},{"id":"https://openalex.org/keywords/focus","display_name":"Focus (optics)","score":0.2572000026702881}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6503000259399414},{"id":"https://openalex.org/C117896860","wikidata":"https://www.wikidata.org/wiki/Q11376","display_name":"Acceleration","level":2,"score":0.5961999893188477},{"id":"https://openalex.org/C149635348","wikidata":"https://www.wikidata.org/wiki/Q193040","display_name":"Embedded system","level":1,"score":0.3043000102043152},{"id":"https://openalex.org/C9390403","wikidata":"https://www.wikidata.org/wiki/Q3966","display_name":"Computer hardware","level":1,"score":0.2874000072479248},{"id":"https://openalex.org/C459310","wikidata":"https://www.wikidata.org/wiki/Q117801","display_name":"Computational science","level":1,"score":0.28139999508857727},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.2696000039577484},{"id":"https://openalex.org/C113843644","wikidata":"https://www.wikidata.org/wiki/Q901882","display_name":"Interface (matter)","level":4,"score":0.2669000029563904},{"id":"https://openalex.org/C121684516","wikidata":"https://www.wikidata.org/wiki/Q7600677","display_name":"Computer graphics (images)","level":1,"score":0.26570001244544983},{"id":"https://openalex.org/C192209626","wikidata":"https://www.wikidata.org/wiki/Q190909","display_name":"Focus (optics)","level":2,"score":0.2572000026702881},{"id":"https://openalex.org/C199639397","wikidata":"https://www.wikidata.org/wiki/Q1788588","display_name":"Engineering drawing","level":1,"score":0.25440001487731934},{"id":"https://openalex.org/C2777904410","wikidata":"https://www.wikidata.org/wiki/Q7397","display_name":"Software","level":2,"score":0.250900000333786}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/hpca68181.2026.11408577","is_oa":false,"landing_page_url":"https://doi.org/10.1109/hpca68181.2026.11408577","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2026 IEEE International Symposium on High Performance Computer Architecture (HPCA)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[{"score":0.8552982807159424,"id":"https://metadata.un.org/sdg/7","display_name":"Affordable and clean energy"}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":21,"referenced_works":["https://openalex.org/W1990653637","https://openalex.org/W1994211684","https://openalex.org/W2024171325","https://openalex.org/W2026930513","https://openalex.org/W2060108852","https://openalex.org/W2107745473","https://openalex.org/W2122962290","https://openalex.org/W2289252105","https://openalex.org/W2945146780","https://openalex.org/W3043522871","https://openalex.org/W3139163275","https://openalex.org/W3209399855","https://openalex.org/W4321636575","https://openalex.org/W4402670692","https://openalex.org/W4404401018","https://openalex.org/W4404781102","https://openalex.org/W4404955001","https://openalex.org/W4411486062","https://openalex.org/W4412945475","https://openalex.org/W4414974080","https://openalex.org/W7103748828"],"related_works":[],"abstract_inverted_index":{"The":[0],"ever-growing":[1],"sizes":[2],"of":[3,40,52,74,82,138,140,174,245],"large":[4,127],"language":[5],"models":[6],"(LLMs)":[7],"introduce":[8],"significant":[9],"infrastructure":[10],"challenges":[11],"due":[12,166],"to":[13,26,48,113,118,133,151,167,199,256,279,290,304],"their":[14],"immense":[15],"memory":[16,35,77,128,144,203,230,300],"capacity":[17,78,129,145,204,231,301],"demands.":[18],"While":[19],"the":[20,37,49,83,93,143,168,195,235,242,274,305],"de":[21],"facto":[22],"approach":[23],"has":[24,43,60,130],"been":[25],"deploy":[27],"multiple":[28],"high-end":[29],"GPUs,":[30],"each":[31],"with":[32,92,136],"a":[33,45,56,68,80,211,268],"limited":[34],"capacity,":[36],"prohibitive":[38],"cost":[39,84],"such":[41],"systems":[42],"become":[44,61,131],"major":[46],"barrier":[47],"widespread":[50],"deployment":[51],"frontier":[53],"LLMs.":[54],"As":[55],"result,":[57],"CPU-based":[58],"inference":[59,122,164,186,201,228,261,286,307],"an":[62,72,184],"appealing":[63],"and":[64,154,171,214,238,254,292,296],"cost-efficient":[65],"alternative,":[66],"since":[67],"CPU":[69,116],"can":[70],"offer":[71],"order":[73],"magnitude":[75],"larger":[76],"at":[79],"fraction":[81],"while":[85,233],"providing":[86],"competitive":[87],"throughput":[88],"for":[89,104,294],"matrixvector":[90],"multiplication":[91],"latest":[94,196],"Advanced":[95,250],"Matrix":[96],"Extensions":[97,252],"(AMX).":[98],"It":[99],"not":[100],"only":[101],"broadens":[102],"accessibility":[103],"users":[105],"without":[106,312],"multiGPU":[107],"setups":[108],"but":[109,159],"also":[110],"enables":[111,222],"hyperscalers":[112],"leverage":[114],"underutilized":[115],"servers":[117],"accommodate":[119],"temporarily":[120],"surging":[121],"demand.":[123],"Nevertheless,":[124],"even":[125],"CPU's":[126],"insufficient":[132],"serve":[134],"LLMs":[135],"hundreds":[137],"billions":[139],"parameters.":[141],"Under":[142],"constraint,":[146],"we":[147,181],"may":[148],"offload":[149],"parameters":[150,209],"storage":[152,175,225],"devices":[153],"fetch":[155],"them":[156,216],"on":[157,217,310],"demand,":[158],"doing":[160],"so":[161],"significantly":[162,223],"degrades":[163],"performance":[165],"high":[169],"latency":[170,287],"low":[172],"bandwidth":[173],"devices.":[176],"To":[177],"address":[178],"this":[179],"challenge,":[180],"propose":[182],"LILO,":[183],"LLM":[185],"framework":[187],"that":[188,272,283],"leverages":[189],"In-memory":[190],"Analytics":[191],"Accelerator":[192],"(IAA)":[193],"in":[194,210],"Intel":[197],"CPUs,":[198],"accelerate":[200],"under":[202,229,299],"constraints.":[205],"By":[206],"storing":[207],"model":[208,236],"compressed":[212],"format":[213],"decompressing":[215],"demand":[218],"using":[219],"IAA,":[220,249],"LILO":[221,240,264,284],"reduced":[224],"access":[226],"during":[227],"constraints":[232,302],"preserving":[234],"accuracy":[237],"behavior.":[239],"orchestrates":[241],"concurrent":[243],"execution":[244],"on-chip":[246],"accelerators,":[247],"i.e.,":[248],"Vector":[251],"(AVX),":[253],"AMX,":[255],"facilitate":[257],"high-throughput":[258],"decompression":[259,275],"alongside":[260],"computation.":[262],"Furthermore,":[263],"implements":[265],"selective":[266],"compression,":[267],"Mixture-of-Expert":[269],"(MoE)-aware":[270],"optimization":[271],"reduces":[273,285],"overhead":[276],"by":[277,288],"up":[278,289],"1.9\u00d7.":[280],"We":[281],"demonstrate":[282],"4.9\u00d7":[291],"4.3\u00d7":[293],"Llama3-405B":[295],"DeepSeekR1,":[297],"respectively,":[298],"compared":[303],"baseline":[306],"solely":[308],"relying":[309],"storage-offloading":[311],"compression.":[313]},"counts_by_year":[],"updated_date":"2026-03-06T06:45:51.903784","created_date":"2026-03-05T00:00:00"}