{"id":"https://openalex.org/W4403534421","doi":"https://doi.org/10.1109/tpami.2024.3483654","title":"Enabling Energy-Efficient Deployment of Large Language Models on Memristor Crossbar: A Synergy of Large and Small","display_name":"Enabling Energy-Efficient Deployment of Large Language Models on Memristor Crossbar: A Synergy of Large and Small","publication_year":2024,"publication_date":"2024-10-18","ids":{"openalex":"https://openalex.org/W4403534421","doi":"https://doi.org/10.1109/tpami.2024.3483654","pmid":"https://pubmed.ncbi.nlm.nih.gov/39423084"},"language":"en","primary_location":{"id":"doi:10.1109/tpami.2024.3483654","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tpami.2024.3483654","pdf_url":null,"source":{"id":"https://openalex.org/S199944782","display_name":"IEEE Transactions on Pattern Analysis and Machine Intelligence","issn_l":"0162-8828","issn":["0162-8828","1939-3539","2160-9292"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310320439","host_organization_name":"IEEE Computer Society","host_organization_lineage":["https://openalex.org/P4310320439","https://openalex.org/P4310319808"],"host_organization_lineage_names":["IEEE Computer Society","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 Transactions on Pattern Analysis and Machine Intelligence","raw_type":"journal-article"},"type":"article","indexed_in":["arxiv","crossref","pubmed"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2410.15977","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5101482096","display_name":"Zhehui Wang","orcid":"https://orcid.org/0000-0002-7139-724X"},"institutions":[{"id":"https://openalex.org/I115228651","display_name":"Agency for Science, Technology and Research","ror":"https://ror.org/036wvzt09","country_code":"SG","type":"government","lineage":["https://openalex.org/I115228651"]},{"id":"https://openalex.org/I3004594783","display_name":"Institute of High Performance Computing","ror":"https://ror.org/02n0ejh50","country_code":"SG","type":"facility","lineage":["https://openalex.org/I115228651","https://openalex.org/I3004594783","https://openalex.org/I91275662"]}],"countries":["SG"],"is_corresponding":true,"raw_author_name":"Zhehui Wang","raw_affiliation_strings":["Institute of High Performance Computing (IHPC), Agency for Science, Technology and Research (A*STAR), Singapore"],"raw_orcid":"https://orcid.org/0000-0002-7139-724X","affiliations":[{"raw_affiliation_string":"Institute of High Performance Computing (IHPC), Agency for Science, Technology and Research (A*STAR), Singapore","institution_ids":["https://openalex.org/I3004594783","https://openalex.org/I115228651"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101630043","display_name":"Tao Luo","orcid":"https://orcid.org/0000-0002-3415-3676"},"institutions":[{"id":"https://openalex.org/I115228651","display_name":"Agency for Science, Technology and Research","ror":"https://ror.org/036wvzt09","country_code":"SG","type":"government","lineage":["https://openalex.org/I115228651"]},{"id":"https://openalex.org/I3004594783","display_name":"Institute of High Performance Computing","ror":"https://ror.org/02n0ejh50","country_code":"SG","type":"facility","lineage":["https://openalex.org/I115228651","https://openalex.org/I3004594783","https://openalex.org/I91275662"]}],"countries":["SG"],"is_corresponding":false,"raw_author_name":"Tao Luo","raw_affiliation_strings":["Institute of High Performance Computing (IHPC), Agency for Science, Technology and Research (A*STAR), Singapore"],"raw_orcid":"https://orcid.org/0000-0002-3415-3676","affiliations":[{"raw_affiliation_string":"Institute of High Performance Computing (IHPC), Agency for Science, Technology and Research (A*STAR), Singapore","institution_ids":["https://openalex.org/I3004594783","https://openalex.org/I115228651"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100384309","display_name":"Cheng Liu","orcid":"https://orcid.org/0000-0002-5542-7306"},"institutions":[{"id":"https://openalex.org/I19820366","display_name":"Chinese Academy of Sciences","ror":"https://ror.org/034t30j35","country_code":"CN","type":"government","lineage":["https://openalex.org/I19820366"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Cheng Liu","raw_affiliation_strings":["Chinese Academy of Sciences, Beijing, China","Chinese Academy of Sciences, China"],"raw_orcid":"https://orcid.org/0000-0002-5542-7306","affiliations":[{"raw_affiliation_string":"Chinese Academy of Sciences, Beijing, China","institution_ids":["https://openalex.org/I19820366"]},{"raw_affiliation_string":"Chinese Academy of Sciences, China","institution_ids":["https://openalex.org/I19820366"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5038598144","display_name":"Weichen Liu","orcid":"https://orcid.org/0000-0001-9348-4662"},"institutions":[{"id":"https://openalex.org/I172675005","display_name":"Nanyang Technological University","ror":"https://ror.org/02e7b5302","country_code":"SG","type":"education","lineage":["https://openalex.org/I172675005"]}],"countries":["SG"],"is_corresponding":false,"raw_author_name":"Weichen Liu","raw_affiliation_strings":["Nanyang Technological University, Singapore"],"raw_orcid":"https://orcid.org/0000-0001-9348-4662","affiliations":[{"raw_affiliation_string":"Nanyang Technological University, Singapore","institution_ids":["https://openalex.org/I172675005"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5001599295","display_name":"Rick Siow Mong Goh","orcid":"https://orcid.org/0000-0001-9116-1595"},"institutions":[{"id":"https://openalex.org/I115228651","display_name":"Agency for Science, Technology and Research","ror":"https://ror.org/036wvzt09","country_code":"SG","type":"government","lineage":["https://openalex.org/I115228651"]},{"id":"https://openalex.org/I3004594783","display_name":"Institute of High Performance Computing","ror":"https://ror.org/02n0ejh50","country_code":"SG","type":"facility","lineage":["https://openalex.org/I115228651","https://openalex.org/I3004594783","https://openalex.org/I91275662"]}],"countries":["SG"],"is_corresponding":false,"raw_author_name":"Rick Siow Mong Goh","raw_affiliation_strings":["Institute of High Performance Computing (IHPC), Agency for Science, Technology and Research (A*STAR), Singapore"],"raw_orcid":"https://orcid.org/0000-0001-9116-1595","affiliations":[{"raw_affiliation_string":"Institute of High Performance Computing (IHPC), Agency for Science, Technology and Research (A*STAR), Singapore","institution_ids":["https://openalex.org/I3004594783","https://openalex.org/I115228651"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5023989495","display_name":"Weng\u2010Fai Wong","orcid":"https://orcid.org/0000-0002-4281-2053"},"institutions":[{"id":"https://openalex.org/I165932596","display_name":"National University of Singapore","ror":"https://ror.org/01tgyzw49","country_code":"SG","type":"education","lineage":["https://openalex.org/I165932596"]}],"countries":["SG"],"is_corresponding":false,"raw_author_name":"Weng-Fai Wong","raw_affiliation_strings":["National University of Singapore, Singapore"],"raw_orcid":"https://orcid.org/0000-0002-4281-2053","affiliations":[{"raw_affiliation_string":"National University of Singapore, Singapore","institution_ids":["https://openalex.org/I165932596"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":6,"corresponding_author_ids":["https://openalex.org/A5101482096"],"corresponding_institution_ids":["https://openalex.org/I115228651","https://openalex.org/I3004594783"],"apc_list":null,"apc_paid":null,"fwci":1.8026,"has_fulltext":true,"cited_by_count":9,"citation_normalized_percentile":{"value":0.856383,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":98,"max":99},"biblio":{"volume":"47","issue":"2","first_page":"916","last_page":"933"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10502","display_name":"Advanced Memory and Neural Computing","score":0.9994000196456909,"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"}},"topics":[{"id":"https://openalex.org/T10502","display_name":"Advanced Memory and Neural Computing","score":0.9994000196456909,"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/T12808","display_name":"Ferroelectric and Negative Capacitance Devices","score":0.9983000159263611,"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/T11948","display_name":"Machine Learning in Materials Science","score":0.9887999892234802,"subfield":{"id":"https://openalex.org/subfields/2505","display_name":"Materials Chemistry"},"field":{"id":"https://openalex.org/fields/25","display_name":"Materials Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/memristor","display_name":"Memristor","score":0.9001013040542603},{"id":"https://openalex.org/keywords/crossbar-switch","display_name":"Crossbar switch","score":0.6743611097335815},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.6511878967285156},{"id":"https://openalex.org/keywords/software-deployment","display_name":"Software deployment","score":0.650336742401123},{"id":"https://openalex.org/keywords/memistor","display_name":"Memistor","score":0.5399231314659119},{"id":"https://openalex.org/keywords/microarchitecture","display_name":"Microarchitecture","score":0.4919526278972626},{"id":"https://openalex.org/keywords/computer-architecture","display_name":"Computer architecture","score":0.36275801062583923},{"id":"https://openalex.org/keywords/resistive-random-access-memory","display_name":"Resistive random-access memory","score":0.3384914994239807},{"id":"https://openalex.org/keywords/embedded-system","display_name":"Embedded system","score":0.27722641825675964},{"id":"https://openalex.org/keywords/telecommunications","display_name":"Telecommunications","score":0.23520734906196594},{"id":"https://openalex.org/keywords/electrical-engineering","display_name":"Electrical engineering","score":0.22485214471817017},{"id":"https://openalex.org/keywords/engineering","display_name":"Engineering","score":0.21387061476707458},{"id":"https://openalex.org/keywords/voltage","display_name":"Voltage","score":0.15677177906036377},{"id":"https://openalex.org/keywords/software-engineering","display_name":"Software engineering","score":0.12137305736541748}],"concepts":[{"id":"https://openalex.org/C150072547","wikidata":"https://www.wikidata.org/wiki/Q212923","display_name":"Memristor","level":2,"score":0.9001013040542603},{"id":"https://openalex.org/C29984679","wikidata":"https://www.wikidata.org/wiki/Q1929149","display_name":"Crossbar switch","level":2,"score":0.6743611097335815},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6511878967285156},{"id":"https://openalex.org/C105339364","wikidata":"https://www.wikidata.org/wiki/Q2297740","display_name":"Software deployment","level":2,"score":0.650336742401123},{"id":"https://openalex.org/C1895703","wikidata":"https://www.wikidata.org/wiki/Q6034938","display_name":"Memistor","level":4,"score":0.5399231314659119},{"id":"https://openalex.org/C107598950","wikidata":"https://www.wikidata.org/wiki/Q259864","display_name":"Microarchitecture","level":2,"score":0.4919526278972626},{"id":"https://openalex.org/C118524514","wikidata":"https://www.wikidata.org/wiki/Q173212","display_name":"Computer architecture","level":1,"score":0.36275801062583923},{"id":"https://openalex.org/C182019814","wikidata":"https://www.wikidata.org/wiki/Q1143830","display_name":"Resistive random-access memory","level":3,"score":0.3384914994239807},{"id":"https://openalex.org/C149635348","wikidata":"https://www.wikidata.org/wiki/Q193040","display_name":"Embedded system","level":1,"score":0.27722641825675964},{"id":"https://openalex.org/C76155785","wikidata":"https://www.wikidata.org/wiki/Q418","display_name":"Telecommunications","level":1,"score":0.23520734906196594},{"id":"https://openalex.org/C119599485","wikidata":"https://www.wikidata.org/wiki/Q43035","display_name":"Electrical engineering","level":1,"score":0.22485214471817017},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.21387061476707458},{"id":"https://openalex.org/C165801399","wikidata":"https://www.wikidata.org/wiki/Q25428","display_name":"Voltage","level":2,"score":0.15677177906036377},{"id":"https://openalex.org/C115903868","wikidata":"https://www.wikidata.org/wiki/Q80993","display_name":"Software engineering","level":1,"score":0.12137305736541748}],"mesh":[],"locations_count":3,"locations":[{"id":"doi:10.1109/tpami.2024.3483654","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tpami.2024.3483654","pdf_url":null,"source":{"id":"https://openalex.org/S199944782","display_name":"IEEE Transactions on Pattern Analysis and Machine Intelligence","issn_l":"0162-8828","issn":["0162-8828","1939-3539","2160-9292"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310320439","host_organization_name":"IEEE Computer Society","host_organization_lineage":["https://openalex.org/P4310320439","https://openalex.org/P4310319808"],"host_organization_lineage_names":["IEEE Computer Society","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 Transactions on Pattern Analysis and Machine Intelligence","raw_type":"journal-article"},{"id":"pmid:39423084","is_oa":false,"landing_page_url":"https://pubmed.ncbi.nlm.nih.gov/39423084","pdf_url":null,"source":{"id":"https://openalex.org/S4306525036","display_name":"PubMed","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I1299303238","host_organization_name":"National Institutes of Health","host_organization_lineage":["https://openalex.org/I1299303238"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE transactions on pattern analysis and machine intelligence","raw_type":null},{"id":"pmh:oai:arXiv.org:2410.15977","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2410.15977","pdf_url":"https://arxiv.org/pdf/2410.15977","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"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":"text"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2410.15977","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2410.15977","pdf_url":"https://arxiv.org/pdf/2410.15977","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"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":"text"},"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/7","score":0.8600000143051147,"display_name":"Affordable and clean energy"}],"awards":[],"funders":[{"id":"https://openalex.org/F4320320751","display_name":"Ministry of Education - Singapore","ror":"https://ror.org/01kcva023"}],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4403534421.pdf","grobid_xml":"https://content.openalex.org/works/W4403534421.grobid-xml"},"referenced_works_count":56,"referenced_works":["https://openalex.org/W48535165","https://openalex.org/W600310561","https://openalex.org/W1878252571","https://openalex.org/W2008311222","https://openalex.org/W2013028205","https://openalex.org/W2047478832","https://openalex.org/W2048266589","https://openalex.org/W2085424870","https://openalex.org/W2105102111","https://openalex.org/W2155589054","https://openalex.org/W2162279286","https://openalex.org/W2164819109","https://openalex.org/W2167221365","https://openalex.org/W2167313300","https://openalex.org/W2170987901","https://openalex.org/W2293423139","https://openalex.org/W2333849942","https://openalex.org/W2399958287","https://openalex.org/W2487018225","https://openalex.org/W2508602506","https://openalex.org/W2518281301","https://openalex.org/W2605663629","https://openalex.org/W2613989746","https://openalex.org/W2761149222","https://openalex.org/W2795127895","https://openalex.org/W2805362231","https://openalex.org/W2892120106","https://openalex.org/W2906472504","https://openalex.org/W2909633668","https://openalex.org/W2923014074","https://openalex.org/W2981124343","https://openalex.org/W3005875844","https://openalex.org/W3006600764","https://openalex.org/W3013080934","https://openalex.org/W3018945530","https://openalex.org/W3033826949","https://openalex.org/W3034855459","https://openalex.org/W3045949243","https://openalex.org/W3091835145","https://openalex.org/W3113256013","https://openalex.org/W3126444928","https://openalex.org/W3134195144","https://openalex.org/W3159868971","https://openalex.org/W3205015346","https://openalex.org/W4200364535","https://openalex.org/W4229035884","https://openalex.org/W4231109836","https://openalex.org/W4237346011","https://openalex.org/W4239972437","https://openalex.org/W4378800995","https://openalex.org/W4380874786","https://openalex.org/W4382203227","https://openalex.org/W4396653761","https://openalex.org/W6638915636","https://openalex.org/W6763189659","https://openalex.org/W6856529530"],"related_works":["https://openalex.org/W2171130799","https://openalex.org/W3173413269","https://openalex.org/W4253441086","https://openalex.org/W3164474614","https://openalex.org/W2548135880","https://openalex.org/W2015477599","https://openalex.org/W2516929886","https://openalex.org/W3177379469","https://openalex.org/W2005875039","https://openalex.org/W2015497999"],"abstract_inverted_index":{"Large":[0],"language":[1],"models":[2],"(LLMs)":[3],"have":[4,37],"garnered":[5],"substantial":[6],"attention":[7,111],"due":[8],"to":[9,63,195,205,214],"their":[10],"promising":[11,41],"applications":[12],"in":[13,27,53,142,206,210,225],"diverse":[14],"domains.":[15],"Nevertheless,":[16],"the":[17,28,74,92,100,160,165,177,226],"increasing":[18],"size":[19,77,93],"of":[20,94,102,137,167,203],"LLMs":[21,83,95,107],"comes":[22],"with":[23,79,183],"a":[24,40,45,156,171,223,230],"significant":[25,231],"surge":[26],"computational":[29],"requirements":[30],"for":[31,71,159],"training":[32],"and":[33,48,147,179,209,229],"deployment.":[34],"Memristor":[35],"crossbars":[36,86,121,127],"emerged":[38],"as":[39,145],"solution,":[42],"which":[43,113],"demonstrated":[44],"small":[46],"footprint":[47],"remarkably":[49],"high":[50],"energy":[51,178,211,233],"efficiency":[52],"computer":[54],"vision":[55],"(CV)":[56],"models.":[57],"Memristors":[58],"possess":[59],"higher":[60],"density":[61],"compared":[62],"conventional":[64],"memory":[65],"technologies,":[66],"making":[67],"them":[68],"highly":[69],"suitable":[70],"effectively":[72],"managing":[73],"extreme":[75],"model":[76],"associated":[78,182],"LLMs.":[80],"However,":[81],"deploying":[82],"on":[84,170,188],"memristor":[85,104,120,126,161,197],"faces":[87],"three":[88],"major":[89],"challenges.":[90],"First,":[91],"increases":[96],"rapidly,":[97],"already":[98],"surpassing":[99],"capabilities":[101],"state-of-the-art":[103,168],"chips.":[105],"Second,":[106],"often":[108],"incorporate":[109],"multi-head":[110],"blocks,":[112],"involve":[114],"non-weight":[115],"stationary":[116],"multiplications":[117],"that":[118,163],"traditional":[119,196],"cannot":[122],"support.":[123],"Third,":[124],"while":[125],"excel":[128],"at":[129,221],"performing":[130],"linear":[131],"operations,":[132],"they":[133],"are":[134],"not":[135],"capable":[136],"executing":[138],"complex":[139],"nonlinear":[140],"operations":[141],"LLM":[143,169],"such":[144],"softmax":[146],"layer":[148],"normalization.":[149],"To":[150],"address":[151],"these":[152],"challenges,":[153],"we":[154],"present":[155],"novel":[157],"architecture":[158,200,219],"crossbar":[162],"enables":[164],"deployment":[166],"single":[172],"chip":[173],"or":[174],"package,":[175],"eliminating":[176],"time":[180],"inefficiencies":[181],"off-chip":[184],"communication.":[185],"Our":[186],"testing":[187],"BERT":[189],"showed":[190],"negligible":[191],"accuracy":[192],"loss.":[193],"Compared":[194,213],"crossbars,":[198],"our":[199,218],"achieves":[201],"enhancements":[202],"up":[204],"area":[207],"overhead":[208],"consumption.":[212],"modern":[215],"TPU/GPU":[216],"systems,":[217],"demonstrates":[220],"least":[222],"reduction":[224],"area-delay":[227],"product":[228],"69%":[232],"consumption":[234],"reduction.":[235]},"counts_by_year":[{"year":2026,"cited_by_count":2},{"year":2025,"cited_by_count":7}],"updated_date":"2026-03-27T05:58:40.876381","created_date":"2025-10-10T00:00:00"}