{"id":"https://openalex.org/W4414197041","doi":"https://doi.org/10.1109/dac63849.2025.11133391","title":"PIMPAL: Accelerating LLM Inference on Edge Devices via In-DRAM Arithmetic Lookup","display_name":"PIMPAL: Accelerating LLM Inference on Edge Devices via In-DRAM Arithmetic Lookup","publication_year":2025,"publication_date":"2025-06-22","ids":{"openalex":"https://openalex.org/W4414197041","doi":"https://doi.org/10.1109/dac63849.2025.11133391"},"language":"en","primary_location":{"id":"doi:10.1109/dac63849.2025.11133391","is_oa":false,"landing_page_url":"https://doi.org/10.1109/dac63849.2025.11133391","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 62nd ACM/IEEE Design Automation Conference (DAC)","raw_type":"proceedings-article"},"type":"conference-paper","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":"Yoonho Jang","orcid":null},"institutions":[{"id":"https://openalex.org/I848706","display_name":"Sungkyunkwan University","ror":"https://ror.org/04q78tk20","country_code":"KR","type":"education","lineage":["https://openalex.org/I848706"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Yoonho Jang","raw_affiliation_strings":["Sungkyunkwan University,Department of Electrical and Computer Engineering,Suwon,Republic of Korea"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Sungkyunkwan University,Department of Electrical and Computer Engineering,Suwon,Republic of Korea","institution_ids":["https://openalex.org/I848706"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Hyeongjun Cho","orcid":null},"institutions":[{"id":"https://openalex.org/I848706","display_name":"Sungkyunkwan University","ror":"https://ror.org/04q78tk20","country_code":"KR","type":"education","lineage":["https://openalex.org/I848706"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Hyeongjun Cho","raw_affiliation_strings":["Sungkyunkwan University,Department of Semiconductor Convergence Engineering,Suwon,Republic of Korea"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Sungkyunkwan University,Department of Semiconductor Convergence Engineering,Suwon,Republic of Korea","institution_ids":["https://openalex.org/I848706"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5047352904","display_name":"Yesin Ryu","orcid":"https://orcid.org/0000-0002-2678-7396"},"institutions":[{"id":"https://openalex.org/I848706","display_name":"Sungkyunkwan University","ror":"https://ror.org/04q78tk20","country_code":"KR","type":"education","lineage":["https://openalex.org/I848706"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Yesin Ryu","raw_affiliation_strings":["Sungkyunkwan University,Department of Semiconductor and Display Engineering,Suwon,Republic of Korea"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Sungkyunkwan University,Department of Semiconductor and Display Engineering,Suwon,Republic of Korea","institution_ids":["https://openalex.org/I848706"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100354952","display_name":"Jung Rae Kim","orcid":"https://orcid.org/0000-0003-0103-7457"},"institutions":[{"id":"https://openalex.org/I848706","display_name":"Sungkyunkwan University","ror":"https://ror.org/04q78tk20","country_code":"KR","type":"education","lineage":["https://openalex.org/I848706"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Jungrae Kim","raw_affiliation_strings":["Sungkyunkwan University,Department of Electrical and Computer Engineering,Suwon,Republic of Korea"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Sungkyunkwan University,Department of Electrical and Computer Engineering,Suwon,Republic of Korea","institution_ids":["https://openalex.org/I848706"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5062176509","display_name":"Seokin Hong","orcid":"https://orcid.org/0000-0001-7842-125X"},"institutions":[{"id":"https://openalex.org/I848706","display_name":"Sungkyunkwan University","ror":"https://ror.org/04q78tk20","country_code":"KR","type":"education","lineage":["https://openalex.org/I848706"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Seokin Hong","raw_affiliation_strings":["Sungkyunkwan University,Department of Electrical and Computer Engineering,Suwon,Republic of Korea"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Sungkyunkwan University,Department of Electrical and Computer Engineering,Suwon,Republic of Korea","institution_ids":["https://openalex.org/I848706"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I848706"],"apc_list":null,"apc_paid":null,"fwci":7.8743,"has_fulltext":false,"cited_by_count":6,"citation_normalized_percentile":{"value":0.97941817,"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":"1","last_page":"7"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.9908999800682068,"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":0.9908999800682068,"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/T11181","display_name":"Advanced Data Storage Technologies","score":0.9907000064849854,"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/T10558","display_name":"Advancements in Semiconductor Devices and Circuit Design","score":0.9628999829292297,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/lookup-table","display_name":"Lookup table","score":0.8019000291824341},{"id":"https://openalex.org/keywords/overhead","display_name":"Overhead (engineering)","score":0.6536999940872192},{"id":"https://openalex.org/keywords/dram","display_name":"Dram","score":0.6424000263214111},{"id":"https://openalex.org/keywords/table","display_name":"Table (database)","score":0.5860999822616577},{"id":"https://openalex.org/keywords/key","display_name":"Key (lock)","score":0.5637999773025513},{"id":"https://openalex.org/keywords/computation","display_name":"Computation","score":0.4659999907016754},{"id":"https://openalex.org/keywords/enhanced-data-rates-for-gsm-evolution","display_name":"Enhanced Data Rates for GSM Evolution","score":0.44029998779296875},{"id":"https://openalex.org/keywords/multiplication","display_name":"Multiplication (music)","score":0.420199990272522},{"id":"https://openalex.org/keywords/edge-device","display_name":"Edge device","score":0.384799987077713}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8057000041007996},{"id":"https://openalex.org/C134835016","wikidata":"https://www.wikidata.org/wiki/Q690265","display_name":"Lookup table","level":2,"score":0.8019000291824341},{"id":"https://openalex.org/C2779960059","wikidata":"https://www.wikidata.org/wiki/Q7113681","display_name":"Overhead (engineering)","level":2,"score":0.6536999940872192},{"id":"https://openalex.org/C7366592","wikidata":"https://www.wikidata.org/wiki/Q1255620","display_name":"Dram","level":2,"score":0.6424000263214111},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.6270999908447266},{"id":"https://openalex.org/C45235069","wikidata":"https://www.wikidata.org/wiki/Q278425","display_name":"Table (database)","level":2,"score":0.5860999822616577},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.5637999773025513},{"id":"https://openalex.org/C45374587","wikidata":"https://www.wikidata.org/wiki/Q12525525","display_name":"Computation","level":2,"score":0.4659999907016754},{"id":"https://openalex.org/C162307627","wikidata":"https://www.wikidata.org/wiki/Q204833","display_name":"Enhanced Data Rates for GSM Evolution","level":2,"score":0.44029998779296875},{"id":"https://openalex.org/C2780595030","wikidata":"https://www.wikidata.org/wiki/Q3860309","display_name":"Multiplication (music)","level":2,"score":0.420199990272522},{"id":"https://openalex.org/C138236772","wikidata":"https://www.wikidata.org/wiki/Q25098575","display_name":"Edge device","level":3,"score":0.384799987077713},{"id":"https://openalex.org/C94375191","wikidata":"https://www.wikidata.org/wiki/Q11205","display_name":"Arithmetic","level":1,"score":0.37439998984336853},{"id":"https://openalex.org/C118524514","wikidata":"https://www.wikidata.org/wiki/Q173212","display_name":"Computer architecture","level":1,"score":0.3668000102043152},{"id":"https://openalex.org/C2778456923","wikidata":"https://www.wikidata.org/wiki/Q5337692","display_name":"Edge computing","level":3,"score":0.335999995470047},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.32749998569488525},{"id":"https://openalex.org/C106515295","wikidata":"https://www.wikidata.org/wiki/Q26806595","display_name":"Parallel processing","level":2,"score":0.31520000100135803},{"id":"https://openalex.org/C179799912","wikidata":"https://www.wikidata.org/wiki/Q205084","display_name":"Computational complexity theory","level":2,"score":0.3127000033855438},{"id":"https://openalex.org/C123657996","wikidata":"https://www.wikidata.org/wiki/Q12271","display_name":"Architecture","level":2,"score":0.3125999867916107},{"id":"https://openalex.org/C9390403","wikidata":"https://www.wikidata.org/wiki/Q3966","display_name":"Computer hardware","level":1,"score":0.2888000011444092},{"id":"https://openalex.org/C182775192","wikidata":"https://www.wikidata.org/wiki/Q913725","display_name":"Saturation arithmetic","level":3,"score":0.2775999903678894},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.27320000529289246},{"id":"https://openalex.org/C152890283","wikidata":"https://www.wikidata.org/wiki/Q4129922","display_name":"Computing with Memory","level":5,"score":0.26910001039505005},{"id":"https://openalex.org/C164620267","wikidata":"https://www.wikidata.org/wiki/Q376953","display_name":"Adder","level":3,"score":0.26460000872612},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.26260000467300415},{"id":"https://openalex.org/C162319229","wikidata":"https://www.wikidata.org/wiki/Q175263","display_name":"Data structure","level":2,"score":0.2606000006198883},{"id":"https://openalex.org/C14580979","wikidata":"https://www.wikidata.org/wiki/Q876049","display_name":"Very-large-scale integration","level":2,"score":0.2558000087738037},{"id":"https://openalex.org/C83283714","wikidata":"https://www.wikidata.org/wiki/Q121117","display_name":"Supercomputer","level":2,"score":0.2558000087738037},{"id":"https://openalex.org/C176649486","wikidata":"https://www.wikidata.org/wiki/Q2308807","display_name":"Memory management","level":3,"score":0.2529999911785126}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/dac63849.2025.11133391","is_oa":false,"landing_page_url":"https://doi.org/10.1109/dac63849.2025.11133391","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 62nd ACM/IEEE Design Automation Conference (DAC)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":22,"referenced_works":["https://openalex.org/W1981943579","https://openalex.org/W2002555321","https://openalex.org/W2034861439","https://openalex.org/W2117310901","https://openalex.org/W2332254524","https://openalex.org/W2607352011","https://openalex.org/W2725159389","https://openalex.org/W2766489088","https://openalex.org/W2982008795","https://openalex.org/W2983617330","https://openalex.org/W3034457371","https://openalex.org/W3100710793","https://openalex.org/W3138893848","https://openalex.org/W3147974890","https://openalex.org/W3170887803","https://openalex.org/W3189166979","https://openalex.org/W4221001402","https://openalex.org/W4243928162","https://openalex.org/W4295036115","https://openalex.org/W4301607055","https://openalex.org/W4312076517","https://openalex.org/W4393407316"],"related_works":[],"abstract_inverted_index":{"Deploying":[0],"Large":[1],"Language":[2],"Models":[3],"(LLMs)":[4],"on":[5],"edge":[6,41],"devices":[7],"poses":[8],"significant":[9],"challenges":[10],"due":[11],"to":[12,37,54,138,173,191],"their":[13],"high":[14,61,111],"computational":[15,66],"and":[16,147,183],"memory":[17],"demands.":[18],"In":[19],"particular,":[20],"General":[21],"MatrixVector":[22],"Multiplication":[23],"(GEMV),":[24],"a":[25,51,78,154],"key":[26,116],"operation":[27],"in":[28,90],"LLM":[29],"inference,":[30],"is":[31],"highly":[32],"memory-intensive,":[33],"making":[34],"it":[35,119,132,149],"difficult":[36],"accelerate":[38],"using":[39],"conventional":[40,192],"computing":[42],"systems.":[43],"While":[44],"Processing-in-memory":[45],"(PIM)":[46],"architectures":[47],"have":[48],"emerged":[49],"as":[50],"promising":[52],"solution":[53],"this":[55],"challenge,":[56],"they":[57],"often":[58],"suffer":[59],"from":[60,162],"area":[62,107,185],"overhead":[63,108,186],"or":[64],"restricted":[65],"precision.":[67],"This":[68],"paper":[69],"proposes":[70],"PIMPAL":[71,104,113,170],"(Processing-In-Memory":[72],"architecture":[73,81],"with":[74,99],"Parallel":[75],"Arithmetic":[76],"Lookup),":[77],"cost-effective":[79],"PIM":[80,181,195],"leveraging":[82],"LookUp":[83],"Table":[84],"(LUT)-based":[85],"computation":[86],"for":[87,127],"GEMV":[88],"acceleration":[89],"sLLMs":[91],"(small":[92],"LLMs).":[93],"By":[94],"replacing":[95],"traditional":[96],"arithmetic":[97],"operations":[98],"parallel":[100,128],"in-DRAM":[101],"LUT":[102,129,144,155],"lookups,":[103],"significantly":[105],"reduces":[106,184],"while":[109],"maintaining":[110],"performance.":[112],"introduces":[114],"three":[115],"innovations:":[117],"(1)":[118],"divides":[120],"DRAM":[121],"bank":[122],"subarrays":[123],"into":[124],"compute":[125],"blocks":[126],"processing;":[130],"(2)":[131],"employs":[133],"Localityaware":[134],"Compute":[135],"Mapping":[136],"(LCM)":[137],"reduce":[139],"row":[140],"activations":[141],"by":[142,187],"maximizing":[143],"access":[145],"locality;":[146],"(3)":[148],"enables":[150],"multi-precision":[151],"computations":[152],"through":[153],"Aggregation":[156],"(LAG)":[157],"mechanism":[158],"that":[159,169],"combines":[160],"results":[161,167],"multiple":[163],"small":[164],"LUTs.":[165],"Experimental":[166],"show":[168],"achieves":[171],"up":[172],"$17.8":[174],"x$":[175],"higher":[176],"performance":[177],"than":[178],"previous":[179],"LUT-based":[180],"designs":[182],"$40":[188],"\\%$":[189],"compared":[190],"processing":[193],"unit-based":[194],"designs.":[196]},"counts_by_year":[{"year":2026,"cited_by_count":5},{"year":2025,"cited_by_count":1}],"updated_date":"2026-08-11T07:18:39.950985","created_date":"2025-10-10T00:00:00"}