{"id":"https://openalex.org/W7117549365","doi":"https://doi.org/10.1109/iccd65941.2025.00086","title":"ASMA: An Anisotropy Scaling Memristor-Based Accelerator for LLM Inference","display_name":"ASMA: An Anisotropy Scaling Memristor-Based Accelerator for LLM Inference","publication_year":2025,"publication_date":"2025-11-10","ids":{"openalex":"https://openalex.org/W7117549365","doi":"https://doi.org/10.1109/iccd65941.2025.00086"},"language":null,"primary_location":{"id":"doi:10.1109/iccd65941.2025.00086","is_oa":false,"landing_page_url":"https://doi.org/10.1109/iccd65941.2025.00086","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 IEEE 43rd International Conference on Computer Design (ICCD)","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":"https://openalex.org/A5076684924","display_name":"Zijian Xiong","orcid":"https://orcid.org/0009-0007-0790-1548"},"institutions":[{"id":"https://openalex.org/I47720641","display_name":"Huazhong University of Science and Technology","ror":"https://ror.org/00p991c53","country_code":"CN","type":"education","lineage":["https://openalex.org/I47720641"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zijian Xiong","raw_affiliation_strings":["Huazhong University of Science and Technology"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Huazhong University of Science and Technology","institution_ids":["https://openalex.org/I47720641"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5121545611","display_name":"Xiangrui Yang","orcid":null},"institutions":[{"id":"https://openalex.org/I47720641","display_name":"Huazhong University of Science and Technology","ror":"https://ror.org/00p991c53","country_code":"CN","type":"education","lineage":["https://openalex.org/I47720641"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xiangrui Yang","raw_affiliation_strings":["Huazhong University of Science and Technology"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Huazhong University of Science and Technology","institution_ids":["https://openalex.org/I47720641"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5121525570","display_name":"Yuhang Zhang","orcid":null},"institutions":[{"id":"https://openalex.org/I47720641","display_name":"Huazhong University of Science and Technology","ror":"https://ror.org/00p991c53","country_code":"CN","type":"education","lineage":["https://openalex.org/I47720641"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yuhang Zhang","raw_affiliation_strings":["Huazhong University of Science and Technology"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Huazhong University of Science and Technology","institution_ids":["https://openalex.org/I47720641"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5113251852","display_name":"Yue Zhou","orcid":"https://orcid.org/0000-0002-2104-4411"},"institutions":[{"id":"https://openalex.org/I14243506","display_name":"Hong Kong Polytechnic University","ror":"https://ror.org/0030zas98","country_code":"HK","type":"education","lineage":["https://openalex.org/I14243506"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Yue Zhou","raw_affiliation_strings":["The Hong Kong Polytechnic University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"The Hong Kong Polytechnic University","institution_ids":["https://openalex.org/I14243506"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5121586241","display_name":"Jianguo Yang","orcid":null},"institutions":[{"id":"https://openalex.org/I4210090209","display_name":"Institute of Microelectronics","ror":"https://ror.org/009rw8n36","country_code":"SG","type":"facility","lineage":["https://openalex.org/I115228651","https://openalex.org/I4210090209","https://openalex.org/I91275662"]}],"countries":["SG"],"is_corresponding":false,"raw_author_name":"Jianguo Yang","raw_affiliation_strings":["Institute of Microelectronics of the Chinese Academy of Science"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Institute of Microelectronics of the Chinese Academy of Science","institution_ids":["https://openalex.org/I4210090209"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5121577402","display_name":"Yaoyu Tao","orcid":null},"institutions":[{"id":"https://openalex.org/I20231570","display_name":"Peking University","ror":"https://ror.org/02v51f717","country_code":"CN","type":"education","lineage":["https://openalex.org/I20231570"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yaoyu Tao","raw_affiliation_strings":["Institute for Artificial Intelligence, Peking University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Institute for Artificial Intelligence, Peking University","institution_ids":["https://openalex.org/I20231570"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5121594355","display_name":"Xiangshui Miao","orcid":null},"institutions":[{"id":"https://openalex.org/I47720641","display_name":"Huazhong University of Science and Technology","ror":"https://ror.org/00p991c53","country_code":"CN","type":"education","lineage":["https://openalex.org/I47720641"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xiangshui Miao","raw_affiliation_strings":["Huazhong University of Science and Technology"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Huazhong University of Science and Technology","institution_ids":["https://openalex.org/I47720641"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5001690837","display_name":"Yu He","orcid":"https://orcid.org/0000-0001-5212-0176"},"institutions":[{"id":"https://openalex.org/I47720641","display_name":"Huazhong University of Science and Technology","ror":"https://ror.org/00p991c53","country_code":"CN","type":"education","lineage":["https://openalex.org/I47720641"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yuhui He","raw_affiliation_strings":["Huazhong University of Science and Technology"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Huazhong University of Science and Technology","institution_ids":["https://openalex.org/I47720641"]}]}],"institutions":[],"countries_distinct_count":3,"institutions_distinct_count":4,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.5887553,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"562","last_page":"565"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T12808","display_name":"Ferroelectric and Negative Capacitance Devices","score":0.30309998989105225,"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/T12808","display_name":"Ferroelectric and Negative Capacitance Devices","score":0.30309998989105225,"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/T10502","display_name":"Advanced Memory and Neural Computing","score":0.2930000126361847,"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/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.06019999831914902,"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/scaling","display_name":"Scaling","score":0.7071999907493591},{"id":"https://openalex.org/keywords/computation","display_name":"Computation","score":0.588699996471405},{"id":"https://openalex.org/keywords/compiler","display_name":"Compiler","score":0.5875999927520752},{"id":"https://openalex.org/keywords/key","display_name":"Key (lock)","score":0.5602999925613403},{"id":"https://openalex.org/keywords/scalability","display_name":"Scalability","score":0.48010000586509705},{"id":"https://openalex.org/keywords/exploit","display_name":"Exploit","score":0.4699000120162964},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.444599986076355},{"id":"https://openalex.org/keywords/latency","display_name":"Latency (audio)","score":0.3984000086784363}],"concepts":[{"id":"https://openalex.org/C99844830","wikidata":"https://www.wikidata.org/wiki/Q102441924","display_name":"Scaling","level":2,"score":0.7071999907493591},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6866999864578247},{"id":"https://openalex.org/C45374587","wikidata":"https://www.wikidata.org/wiki/Q12525525","display_name":"Computation","level":2,"score":0.588699996471405},{"id":"https://openalex.org/C169590947","wikidata":"https://www.wikidata.org/wiki/Q47506","display_name":"Compiler","level":2,"score":0.5875999927520752},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.5602999925613403},{"id":"https://openalex.org/C48044578","wikidata":"https://www.wikidata.org/wiki/Q727490","display_name":"Scalability","level":2,"score":0.48010000586509705},{"id":"https://openalex.org/C165696696","wikidata":"https://www.wikidata.org/wiki/Q11287","display_name":"Exploit","level":2,"score":0.4699000120162964},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.444599986076355},{"id":"https://openalex.org/C82876162","wikidata":"https://www.wikidata.org/wiki/Q17096504","display_name":"Latency (audio)","level":2,"score":0.3984000086784363},{"id":"https://openalex.org/C85725439","wikidata":"https://www.wikidata.org/wiki/Q466686","display_name":"Anisotropy","level":2,"score":0.37049999833106995},{"id":"https://openalex.org/C31170391","wikidata":"https://www.wikidata.org/wiki/Q188619","display_name":"Hierarchy","level":2,"score":0.3668999969959259},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.3531999886035919},{"id":"https://openalex.org/C2778476105","wikidata":"https://www.wikidata.org/wiki/Q628539","display_name":"Workload","level":2,"score":0.3416000008583069},{"id":"https://openalex.org/C184050105","wikidata":"https://www.wikidata.org/wiki/Q273163","display_name":"Isotropy","level":2,"score":0.3305000066757202},{"id":"https://openalex.org/C2778100165","wikidata":"https://www.wikidata.org/wiki/Q1589327","display_name":"Memory hierarchy","level":3,"score":0.322299987077713},{"id":"https://openalex.org/C118524514","wikidata":"https://www.wikidata.org/wiki/Q173212","display_name":"Computer architecture","level":1,"score":0.2953999936580658},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.2939999997615814},{"id":"https://openalex.org/C2777735758","wikidata":"https://www.wikidata.org/wiki/Q817765","display_name":"Path (computing)","level":2,"score":0.2935999929904938},{"id":"https://openalex.org/C459310","wikidata":"https://www.wikidata.org/wiki/Q117801","display_name":"Computational science","level":1,"score":0.28859999775886536},{"id":"https://openalex.org/C186370098","wikidata":"https://www.wikidata.org/wiki/Q442787","display_name":"Energy (signal processing)","level":2,"score":0.2858999967575073},{"id":"https://openalex.org/C24326235","wikidata":"https://www.wikidata.org/wiki/Q126095","display_name":"Electronic engineering","level":1,"score":0.2806999981403351},{"id":"https://openalex.org/C7366592","wikidata":"https://www.wikidata.org/wiki/Q1255620","display_name":"Dram","level":2,"score":0.2800000011920929},{"id":"https://openalex.org/C2742236","wikidata":"https://www.wikidata.org/wiki/Q924713","display_name":"Efficient energy use","level":2,"score":0.2648000121116638},{"id":"https://openalex.org/C41608201","wikidata":"https://www.wikidata.org/wiki/Q980509","display_name":"Embedding","level":2,"score":0.26420000195503235},{"id":"https://openalex.org/C77618280","wikidata":"https://www.wikidata.org/wiki/Q1155772","display_name":"Scheme (mathematics)","level":2,"score":0.2606000006198883},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.2533000111579895},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.25200000405311584},{"id":"https://openalex.org/C169822122","wikidata":"https://www.wikidata.org/wiki/Q230187","display_name":"Crosstalk","level":2,"score":0.250900000333786}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/iccd65941.2025.00086","is_oa":false,"landing_page_url":"https://doi.org/10.1109/iccd65941.2025.00086","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 IEEE 43rd International Conference on Computer Design (ICCD)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[{"score":0.6654146313667297,"id":"https://metadata.un.org/sdg/7","display_name":"Affordable and clean energy"}],"awards":[{"id":"https://openalex.org/G616102654","display_name":null,"funder_award_id":"2023YFB4502200","funder_id":"https://openalex.org/F4320335777","funder_display_name":"National Key Research and Development Program of China"}],"funders":[{"id":"https://openalex.org/F4320335777","display_name":"National Key Research and Development Program of China","ror":null}],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":16,"referenced_works":["https://openalex.org/W2112181056","https://openalex.org/W2151174323","https://openalex.org/W2508602506","https://openalex.org/W2913104037","https://openalex.org/W3083443371","https://openalex.org/W3209892756","https://openalex.org/W4247806563","https://openalex.org/W4308083827","https://openalex.org/W4385245566","https://openalex.org/W4386763580","https://openalex.org/W4390017976","https://openalex.org/W4390189088","https://openalex.org/W4392367648","https://openalex.org/W4395049340","https://openalex.org/W4404955085","https://openalex.org/W4404955734"],"related_works":[],"abstract_inverted_index":{"Large":[0],"Language":[1],"Models":[2],"(LLMs)":[3],"present":[4],"substantial":[5],"computational":[6],"and":[7,67,81,104,107,134],"memory":[8],"challenges,":[9],"particularly":[10],"within":[11],"their":[12,45],"Feed-Forward":[13],"Network":[14],"(FFN)":[15],"layers.":[16],"While":[17],"Computing-in-Memory":[18],"(CIM)":[19],"using":[20,121],"Memristor":[21],"offers":[22,147],"a":[23,87,94,108,122,148],"path":[24],"to":[25,38,44,132,138,141],"mitigate":[26],"data":[27],"movement":[28],"bottlenecks,":[29],"naively":[30],"scaling":[31,72],"existing":[32],"CIM":[33,144,154],"architectures":[34],"for":[35,61,91,102,151],"LLMs":[36],"leads":[37],"new,":[39],"severe":[40],"communication":[41],"overheads":[42],"due":[43],"typically":[46],"isotropic":[47],"(uniform)":[48],"design.":[49],"This":[50],"paper":[51],"introduces":[52],"ASMA,":[53],"an":[54],"Anisotropy":[55],"Scaling":[56],"Memristor-based":[57],"Accelerator,":[58],"specifically":[59],"co-designed":[60,109],"efficient":[62],"LLM":[63,75,113],"inference.":[64],"ASMA":[65,127,146],"identifies":[66],"exploits":[68],"the":[69],"inherent":[70],"anisotropic":[71,96],"characteristics":[73],"of":[74],"FFN":[76],"layers\u2014categorized":[77],"into":[78],"additive,":[79],"concatenative,":[80],"sequential":[82],"dimensions.":[83],"Key":[84],"innovations":[85],"include":[86],"novel":[88],"Subtile":[89],"hierarchy":[90],"optimized":[92],"pipelining,":[93],"hierarchical":[95,143],"Network-on-Chip":[97],"(NoC)":[98],"featuring":[99],"distinct":[100],"channels":[101],"broadcast":[103],"localized":[105],"communication,":[106],"compiler":[110],"that":[111,126],"maps":[112],"computations":[114],"by":[115,156],"leveraging":[116],"these":[117],"architectural":[118],"specializations.":[119],"Evaluations":[120],"transaction-level":[123],"simulator":[124],"demonstrate":[125],"achieves":[128],"latency":[129],"improvement":[130,136],"up":[131,137],"81.7%":[133],"energy":[135],"91.5%":[139],"compared":[140],"conventional":[142],"designs.":[145],"new":[149],"paradigm":[150],"designing":[152],"LLM-specific":[153],"accelerators":[155],"embracing":[157],"workload":[158],"anisotropy.":[159]},"counts_by_year":[],"updated_date":"2026-07-29T14:22:42.915294","created_date":"2025-12-30T00:00:00"}