{"id":"https://openalex.org/W7084114452","doi":"https://doi.org/10.1109/aicas64808.2025.11173152","title":"Hardware-Aware Optimization of Large Language Models for Enhanced Throughput on Arm CPUs","display_name":"Hardware-Aware Optimization of Large Language Models for Enhanced Throughput on Arm CPUs","publication_year":2025,"publication_date":"2025-04-28","ids":{"openalex":"https://openalex.org/W7084114452","doi":"https://doi.org/10.1109/aicas64808.2025.11173152"},"language":"en","primary_location":{"id":"doi:10.1109/aicas64808.2025.11173152","is_oa":false,"landing_page_url":"https://doi.org/10.1109/aicas64808.2025.11173152","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 IEEE 7th International Conference on Artificial Intelligence Circuits and Systems (AICAS)","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":null,"display_name":"Xingyu Zhu","orcid":null},"institutions":[{"id":"https://openalex.org/I149594827","display_name":"Xidian University","ror":"https://ror.org/05s92vm98","country_code":"CN","type":"education","lineage":["https://openalex.org/I149594827"]}],"countries":["CN"],"is_corresponding":true,"raw_author_name":"Xingyu Zhu","raw_affiliation_strings":["Xidian University,Hangzhou Institute of Technology,Hangzhou,China"],"affiliations":[{"raw_affiliation_string":"Xidian University,Hangzhou Institute of Technology,Hangzhou,China","institution_ids":["https://openalex.org/I149594827"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Longhao Chen","orcid":null},"institutions":[{"id":"https://openalex.org/I50760025","display_name":"Hangzhou Dianzi University","ror":"https://ror.org/0576gt767","country_code":"CN","type":"education","lineage":["https://openalex.org/I50760025"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Longhao Chen","raw_affiliation_strings":["Hangzhou Dianzi University,Zhuoyue Honors College,Hangzhou,China"],"affiliations":[{"raw_affiliation_string":"Hangzhou Dianzi University,Zhuoyue Honors College,Hangzhou,China","institution_ids":["https://openalex.org/I50760025"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Cheng Zhang","orcid":null},"institutions":[{"id":"https://openalex.org/I149594827","display_name":"Xidian University","ror":"https://ror.org/05s92vm98","country_code":"CN","type":"education","lineage":["https://openalex.org/I149594827"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Cheng Zhang","raw_affiliation_strings":["Xidian University,Hangzhou Institute of Technology,Hangzhou,China"],"affiliations":[{"raw_affiliation_string":"Xidian University,Hangzhou Institute of Technology,Hangzhou,China","institution_ids":["https://openalex.org/I149594827"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Pengcheng Yang","orcid":null},"institutions":[{"id":"https://openalex.org/I149594827","display_name":"Xidian University","ror":"https://ror.org/05s92vm98","country_code":"CN","type":"education","lineage":["https://openalex.org/I149594827"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Pengcheng Yang","raw_affiliation_strings":["Xidian University,Hangzhou Institute of Technology,Hangzhou,China"],"affiliations":[{"raw_affiliation_string":"Xidian University,Hangzhou Institute of Technology,Hangzhou,China","institution_ids":["https://openalex.org/I149594827"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Tingjie Yang","orcid":null},"institutions":[{"id":"https://openalex.org/I149594827","display_name":"Xidian University","ror":"https://ror.org/05s92vm98","country_code":"CN","type":"education","lineage":["https://openalex.org/I149594827"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Tingjie Yang","raw_affiliation_strings":["Xidian University,Hangzhou Institute of Technology,Hangzhou,China"],"affiliations":[{"raw_affiliation_string":"Xidian University,Hangzhou Institute of Technology,Hangzhou,China","institution_ids":["https://openalex.org/I149594827"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Guosheng Yu","orcid":null},"institutions":[{"id":"https://openalex.org/I4210119559","display_name":"Taiwan Semiconductor Manufacturing Company (China)","ror":"https://ror.org/02s0wcj29","country_code":"CN","type":"company","lineage":["https://openalex.org/I4210119559","https://openalex.org/I4210120917"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Guosheng Yu","raw_affiliation_strings":["T-HEAD Semiconductor Co., Ltd,China"],"affiliations":[{"raw_affiliation_string":"T-HEAD Semiconductor Co., Ltd,China","institution_ids":["https://openalex.org/I4210119559"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Evens Pan","orcid":null},"institutions":[{"id":"https://openalex.org/I4210156213","display_name":"American Rock Mechanics Association","ror":"https://ror.org/05vfrxy92","country_code":"US","type":"nonprofit","lineage":["https://openalex.org/I4210156213"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Evens Pan","raw_affiliation_strings":["Arm Technology (China) Co., Ltd,China"],"affiliations":[{"raw_affiliation_string":"Arm Technology (China) Co., Ltd,China","institution_ids":["https://openalex.org/I4210156213"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Xiguang Wu","orcid":null},"institutions":[{"id":"https://openalex.org/I149594827","display_name":"Xidian University","ror":"https://ror.org/05s92vm98","country_code":"CN","type":"education","lineage":["https://openalex.org/I149594827"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xiguang Wu","raw_affiliation_strings":["Xidian University,Hangzhou Institute of Technology,Hangzhou,China"],"affiliations":[{"raw_affiliation_string":"Xidian University,Hangzhou Institute of Technology,Hangzhou,China","institution_ids":["https://openalex.org/I149594827"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Bo Li","orcid":null},"institutions":[{"id":"https://openalex.org/I149594827","display_name":"Xidian University","ror":"https://ror.org/05s92vm98","country_code":"CN","type":"education","lineage":["https://openalex.org/I149594827"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Bo Li","raw_affiliation_strings":["Xidian University,Hangzhou Institute of Technology,Hangzhou,China"],"affiliations":[{"raw_affiliation_string":"Xidian University,Hangzhou Institute of Technology,Hangzhou,China","institution_ids":["https://openalex.org/I149594827"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Wei Mao","orcid":null},"institutions":[{"id":"https://openalex.org/I149594827","display_name":"Xidian University","ror":"https://ror.org/05s92vm98","country_code":"CN","type":"education","lineage":["https://openalex.org/I149594827"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Wei Mao","raw_affiliation_strings":["Xidian University,Hangzhou Institute of Technology,Hangzhou,China"],"affiliations":[{"raw_affiliation_string":"Xidian University,Hangzhou Institute of Technology,Hangzhou,China","institution_ids":["https://openalex.org/I149594827"]}]},{"author_position":"last","author":{"id":null,"display_name":"Genquan Han","orcid":null},"institutions":[{"id":"https://openalex.org/I149594827","display_name":"Xidian University","ror":"https://ror.org/05s92vm98","country_code":"CN","type":"education","lineage":["https://openalex.org/I149594827"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Genquan Han","raw_affiliation_strings":["Xidian University,Hangzhou Institute of Technology,Hangzhou,China"],"affiliations":[{"raw_affiliation_string":"Xidian University,Hangzhou Institute of Technology,Hangzhou,China","institution_ids":["https://openalex.org/I149594827"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":11,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I149594827"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.66364949,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"5"},"is_retracted":false,"is_paratext":false,"is_xpac":true,"primary_topic":{"id":"https://openalex.org/T11247","display_name":"Plant nutrient uptake and metabolism","score":0.057999998331069946,"subfield":{"id":"https://openalex.org/subfields/1110","display_name":"Plant Science"},"field":{"id":"https://openalex.org/fields/11","display_name":"Agricultural and Biological Sciences"},"domain":{"id":"https://openalex.org/domains/1","display_name":"Life Sciences"}},"topics":[{"id":"https://openalex.org/T11247","display_name":"Plant nutrient uptake and metabolism","score":0.057999998331069946,"subfield":{"id":"https://openalex.org/subfields/1110","display_name":"Plant Science"},"field":{"id":"https://openalex.org/fields/11","display_name":"Agricultural and Biological Sciences"},"domain":{"id":"https://openalex.org/domains/1","display_name":"Life Sciences"}},{"id":"https://openalex.org/T10184","display_name":"Plant Molecular Biology Research","score":0.0502999983727932,"subfield":{"id":"https://openalex.org/subfields/1110","display_name":"Plant Science"},"field":{"id":"https://openalex.org/fields/11","display_name":"Agricultural and Biological Sciences"},"domain":{"id":"https://openalex.org/domains/1","display_name":"Life Sciences"}},{"id":"https://openalex.org/T12859","display_name":"Cell Image Analysis Techniques","score":0.04610000178217888,"subfield":{"id":"https://openalex.org/subfields/1304","display_name":"Biophysics"},"field":{"id":"https://openalex.org/fields/13","display_name":"Biochemistry, Genetics and Molecular Biology"},"domain":{"id":"https://openalex.org/domains/1","display_name":"Life Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/throughput","display_name":"Throughput","score":0.6825000047683716},{"id":"https://openalex.org/keywords/quantization","display_name":"Quantization (signal processing)","score":0.5238999724388123},{"id":"https://openalex.org/keywords/edge-device","display_name":"Edge device","score":0.5194000005722046},{"id":"https://openalex.org/keywords/process","display_name":"Process (computing)","score":0.5092999935150146},{"id":"https://openalex.org/keywords/software-deployment","display_name":"Software deployment","score":0.4537999927997589},{"id":"https://openalex.org/keywords/computation","display_name":"Computation","score":0.40139999985694885},{"id":"https://openalex.org/keywords/enhanced-data-rates-for-gsm-evolution","display_name":"Enhanced Data Rates for GSM Evolution","score":0.3905999958515167}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8122000098228455},{"id":"https://openalex.org/C157764524","wikidata":"https://www.wikidata.org/wiki/Q1383412","display_name":"Throughput","level":3,"score":0.6825000047683716},{"id":"https://openalex.org/C28855332","wikidata":"https://www.wikidata.org/wiki/Q198099","display_name":"Quantization (signal processing)","level":2,"score":0.5238999724388123},{"id":"https://openalex.org/C138236772","wikidata":"https://www.wikidata.org/wiki/Q25098575","display_name":"Edge device","level":3,"score":0.5194000005722046},{"id":"https://openalex.org/C98045186","wikidata":"https://www.wikidata.org/wiki/Q205663","display_name":"Process (computing)","level":2,"score":0.5092999935150146},{"id":"https://openalex.org/C105339364","wikidata":"https://www.wikidata.org/wiki/Q2297740","display_name":"Software deployment","level":2,"score":0.4537999927997589},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.40799999237060547},{"id":"https://openalex.org/C45374587","wikidata":"https://www.wikidata.org/wiki/Q12525525","display_name":"Computation","level":2,"score":0.40139999985694885},{"id":"https://openalex.org/C162307627","wikidata":"https://www.wikidata.org/wiki/Q204833","display_name":"Enhanced Data Rates for GSM Evolution","level":2,"score":0.3905999958515167},{"id":"https://openalex.org/C137836250","wikidata":"https://www.wikidata.org/wiki/Q984063","display_name":"Optimization problem","level":2,"score":0.38199999928474426},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.3646000027656555},{"id":"https://openalex.org/C18762648","wikidata":"https://www.wikidata.org/wiki/Q42213","display_name":"Work (physics)","level":2,"score":0.34299999475479126},{"id":"https://openalex.org/C2778755073","wikidata":"https://www.wikidata.org/wiki/Q10858537","display_name":"Scale (ratio)","level":2,"score":0.33399999141693115},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.311599999666214},{"id":"https://openalex.org/C66024118","wikidata":"https://www.wikidata.org/wiki/Q1122506","display_name":"Computational model","level":2,"score":0.29670000076293945},{"id":"https://openalex.org/C9390403","wikidata":"https://www.wikidata.org/wiki/Q3966","display_name":"Computer hardware","level":1,"score":0.2922999858856201},{"id":"https://openalex.org/C149635348","wikidata":"https://www.wikidata.org/wiki/Q193040","display_name":"Embedded system","level":1,"score":0.2919999957084656},{"id":"https://openalex.org/C118524514","wikidata":"https://www.wikidata.org/wiki/Q173212","display_name":"Computer architecture","level":1,"score":0.2863999903202057},{"id":"https://openalex.org/C2983448237","wikidata":"https://www.wikidata.org/wiki/Q1078276","display_name":"Language understanding","level":2,"score":0.27639999985694885},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.27410000562667847},{"id":"https://openalex.org/C79403827","wikidata":"https://www.wikidata.org/wiki/Q3988","display_name":"Real-time computing","level":1,"score":0.2734000086784363}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/aicas64808.2025.11173152","is_oa":false,"landing_page_url":"https://doi.org/10.1109/aicas64808.2025.11173152","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 IEEE 7th International Conference on Artificial Intelligence Circuits and Systems (AICAS)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[{"display_name":"Affordable and clean energy","id":"https://metadata.un.org/sdg/7","score":0.6322512030601501}],"awards":[],"funders":[{"id":"https://openalex.org/F4320323230","display_name":"Xidian University","ror":"https://ror.org/05s92vm98"},{"id":"https://openalex.org/F4320337504","display_name":"Research and Development","ror":"https://ror.org/027s68j25"}],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":9,"referenced_works":["https://openalex.org/W2998617917","https://openalex.org/W3012561096","https://openalex.org/W3189877953","https://openalex.org/W4310936523","https://openalex.org/W4312196615","https://openalex.org/W4383108457","https://openalex.org/W4384705353","https://openalex.org/W4385245566","https://openalex.org/W4387321091"],"related_works":[],"abstract_inverted_index":{"Large":[0],"Language":[1],"Models":[2],"(LLMs)":[3],"have":[4],"recently":[5],"demonstrated":[6],"versatile":[7],"capabilities,":[8],"approaching":[9],"artificial":[10],"general":[11],"intelligence":[12],"(AGI).":[13],"The":[14,102,146],"increasing":[15],"scale":[16],"of":[17,56,153],"model":[18,112],"parameters":[19],"necessitates":[20],"high-performance":[21],"computing":[22],"power,":[23],"posing":[24],"a":[25,64],"challenge":[26],"for":[27,79,87,134,142],"efficient":[28],"deployment":[29],"within":[30],"hardware":[31],"limitations.":[32],"For":[33],"mainstream":[34],"edge":[35],"AI":[36],"operation,":[37],"it":[38],"mainly":[39],"involves":[40],"using":[41],"Arm":[42,116],"CPUs":[43],"to":[44,76,120],"perform":[45],"LLM":[46],"inference.":[47],"Thus,":[48],"this":[49,124],"work":[50,104],"concentrates":[51],"on":[52,113],"enhancing":[53],"the":[54,57,91,95,99,109,114,121,135,143],"throughput":[55,128],"prefill":[58,136],"and":[59,138],"decode":[60,144],"phase":[61,137],"by":[62,107],"proposing":[63],"novel":[65],"comprehensive":[66],"optimization":[67,125],"framework.":[68],"A":[69],"mixed-precision":[70],"quantization":[71],"technique":[72],"compresses":[73],"non-critical":[74],"weights":[75,85],"low":[77],"precision":[78],"parallel":[80],"processing,":[81],"while":[82],"keeping":[83],"critical":[84],"high":[86],"accuracy.":[88],"Additionally,":[89],"integrating":[90],"computational":[92,100],"process":[93],"in":[94],"attention":[96],"mechanism":[97],"reduces":[98],"load.":[101],"proposed":[103],"is":[105],"verified":[106],"optimizing":[108],"Qwen":[110],"1.8B":[111],"YiTian":[115],"CPU":[117],"platform.":[118],"Compared":[119],"original":[122],"model,":[123],"achieves":[126],"excellent":[127],"performance":[129],"with":[130],"3.62":[131],"times":[132,140],"higher":[133,141],"13.38":[139],"phase.":[145],"PIQA":[147],"test":[148],"maintains":[149],"an":[150],"accuracy":[151],"rate":[152],"72.52%.":[154]},"counts_by_year":[],"updated_date":"2026-04-09T08:11:56.329763","created_date":"2025-10-10T00:00:00"}