{"id":"https://openalex.org/W4405794828","doi":"https://doi.org/10.1145/3700410.3702126","title":"MNN-LLM: A Generic Inference Engine for Fast Large Language Model Deployment on Mobile Devices","display_name":"MNN-LLM: A Generic Inference Engine for Fast Large Language Model Deployment on Mobile Devices","publication_year":2024,"publication_date":"2024-12-03","ids":{"openalex":"https://openalex.org/W4405794828","doi":"https://doi.org/10.1145/3700410.3702126"},"language":"en","primary_location":{"id":"doi:10.1145/3700410.3702126","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3700410.3702126","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 6th ACM International Conference on Multimedia in Asia Workshops","raw_type":"proceedings-article"},"type":"preprint","indexed_in":["arxiv","crossref"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2506.10443","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5055434702","display_name":"Zhaode Wang","orcid":null},"institutions":[{"id":"https://openalex.org/I45928872","display_name":"Alibaba Group (China)","ror":"https://ror.org/00k642b80","country_code":"CN","type":"company","lineage":["https://openalex.org/I45928872"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zhaode Wang","raw_affiliation_strings":["Alibaba Group, Beijing, China"],"raw_orcid":"https://orcid.org/0009-0009-0498-7491","affiliations":[{"raw_affiliation_string":"Alibaba Group, Beijing, China","institution_ids":["https://openalex.org/I45928872"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Jingbang Yang","orcid":"https://orcid.org/0009-0005-4331-4100"},"institutions":[{"id":"https://openalex.org/I45928872","display_name":"Alibaba Group (China)","ror":"https://ror.org/00k642b80","country_code":"CN","type":"company","lineage":["https://openalex.org/I45928872"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Jingbang Yang","raw_affiliation_strings":["Alibaba Group, Hangzhou, China"],"raw_orcid":"https://orcid.org/0009-0005-4331-4100","affiliations":[{"raw_affiliation_string":"Alibaba Group, Hangzhou, China","institution_ids":["https://openalex.org/I45928872"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5077290644","display_name":"Xinyu Qian","orcid":"https://orcid.org/0009-0009-1687-9746"},"institutions":[{"id":"https://openalex.org/I45928872","display_name":"Alibaba Group (China)","ror":"https://ror.org/00k642b80","country_code":"CN","type":"company","lineage":["https://openalex.org/I45928872"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xinyu Qian","raw_affiliation_strings":["Alibaba Group, Hangzhou, China"],"raw_orcid":"https://orcid.org/0009-0009-1687-9746","affiliations":[{"raw_affiliation_string":"Alibaba Group, Hangzhou, China","institution_ids":["https://openalex.org/I45928872"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5115654520","display_name":"Shiwen Xing","orcid":null},"institutions":[{"id":"https://openalex.org/I45928872","display_name":"Alibaba Group (China)","ror":"https://ror.org/00k642b80","country_code":"CN","type":"company","lineage":["https://openalex.org/I45928872"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Shiwen Xing","raw_affiliation_strings":["Alibaba Group, Hangzhou, China"],"raw_orcid":"https://orcid.org/0009-0005-1353-687X","affiliations":[{"raw_affiliation_string":"Alibaba Group, Hangzhou, China","institution_ids":["https://openalex.org/I45928872"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Xiaotang Jiang","orcid":"https://orcid.org/0009-0006-8995-9841"},"institutions":[{"id":"https://openalex.org/I45928872","display_name":"Alibaba Group (China)","ror":"https://ror.org/00k642b80","country_code":"CN","type":"company","lineage":["https://openalex.org/I45928872"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xiaotang Jiang","raw_affiliation_strings":["Alibaba Group, Hangzhou, China"],"raw_orcid":"https://orcid.org/0009-0006-8995-9841","affiliations":[{"raw_affiliation_string":"Alibaba Group, Hangzhou, China","institution_ids":["https://openalex.org/I45928872"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5020976131","display_name":"Chengfei Lv","orcid":"https://orcid.org/0009-0000-4918-7425"},"institutions":[{"id":"https://openalex.org/I45928872","display_name":"Alibaba Group (China)","ror":"https://ror.org/00k642b80","country_code":"CN","type":"company","lineage":["https://openalex.org/I45928872"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Chengfei Lv","raw_affiliation_strings":["Alibaba Group, Hangzhou, China"],"raw_orcid":"https://orcid.org/0009-0000-4918-7425","affiliations":[{"raw_affiliation_string":"Alibaba Group, Hangzhou, China","institution_ids":["https://openalex.org/I45928872"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5100757087","display_name":"Shengyu Zhang","orcid":"https://orcid.org/0000-0002-0030-8289"},"institutions":[{"id":"https://openalex.org/I76130692","display_name":"Zhejiang University","ror":"https://ror.org/00a2xv884","country_code":"CN","type":"education","lineage":["https://openalex.org/I76130692"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Shengyu Zhang","raw_affiliation_strings":["Zhejiang University, Hangzhou, China"],"raw_orcid":"https://orcid.org/0000-0002-0030-8289","affiliations":[{"raw_affiliation_string":"Zhejiang University, Hangzhou, China","institution_ids":["https://openalex.org/I76130692"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":7,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":2.7371,"has_fulltext":false,"cited_by_count":9,"citation_normalized_percentile":{"value":0.91813759,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":97,"max":100},"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/T10028","display_name":"Topic Modeling","score":0.9259999990463257,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"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/T10028","display_name":"Topic Modeling","score":0.9259999990463257,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"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/software-deployment","display_name":"Software deployment","score":0.7785872220993042},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.7371021509170532},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.7299237847328186},{"id":"https://openalex.org/keywords/inference-engine","display_name":"Inference engine","score":0.5877580046653748},{"id":"https://openalex.org/keywords/language-model","display_name":"Language model","score":0.46389368176460266},{"id":"https://openalex.org/keywords/mobile-device","display_name":"Mobile device","score":0.4217592775821686},{"id":"https://openalex.org/keywords/programming-language","display_name":"Programming language","score":0.32674163579940796},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.2517944574356079},{"id":"https://openalex.org/keywords/software-engineering","display_name":"Software engineering","score":0.18558615446090698},{"id":"https://openalex.org/keywords/operating-system","display_name":"Operating system","score":0.13455575704574585}],"concepts":[{"id":"https://openalex.org/C105339364","wikidata":"https://www.wikidata.org/wiki/Q2297740","display_name":"Software deployment","level":2,"score":0.7785872220993042},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7371021509170532},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.7299237847328186},{"id":"https://openalex.org/C46743427","wikidata":"https://www.wikidata.org/wiki/Q1341685","display_name":"Inference engine","level":3,"score":0.5877580046653748},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.46389368176460266},{"id":"https://openalex.org/C186967261","wikidata":"https://www.wikidata.org/wiki/Q5082128","display_name":"Mobile device","level":2,"score":0.4217592775821686},{"id":"https://openalex.org/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.32674163579940796},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.2517944574356079},{"id":"https://openalex.org/C115903868","wikidata":"https://www.wikidata.org/wiki/Q80993","display_name":"Software engineering","level":1,"score":0.18558615446090698},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.13455575704574585}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1145/3700410.3702126","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3700410.3702126","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 6th ACM International Conference on Multimedia in Asia Workshops","raw_type":"proceedings-article"},{"id":"pmh:oai:arXiv.org:2506.10443","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2506.10443","pdf_url":"https://arxiv.org/pdf/2506.10443","source":{"id":"https://openalex.org/S4393918464","display_name":"ArXiv.org","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":null,"raw_type":"text"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2506.10443","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2506.10443","pdf_url":"https://arxiv.org/pdf/2506.10443","source":{"id":"https://openalex.org/S4393918464","display_name":"ArXiv.org","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":1,"referenced_works":["https://openalex.org/W2790484902"],"related_works":["https://openalex.org/W2057057690","https://openalex.org/W2368184788","https://openalex.org/W2358964818","https://openalex.org/W2359535128","https://openalex.org/W2381332051","https://openalex.org/W2321443665","https://openalex.org/W2375699995","https://openalex.org/W2043719711","https://openalex.org/W2364072231","https://openalex.org/W2140069467"],"abstract_inverted_index":{"Large":[0],"language":[1,63],"models":[2,64],"(LLMs)":[3],"have":[4],"demonstrated":[5],"exceptional":[6],"performance":[7],"across":[8],"a":[9,34,53,122],"variety":[10],"of":[11,40,61,73],"tasks.":[12],"However,":[13],"their":[14],"substantial":[15],"scale":[16],"leads":[17],"to":[18,57,114,121,127],"significant":[19],"computational":[20],"resource":[21],"consumption":[22],"during":[23],"inference,":[24],"resulting":[25],"in":[26],"high":[27],"costs.":[28],"Consequently,":[29],"edge":[30,41],"device":[31],"inference":[32,42,47],"presents":[33],"promising":[35],"solution.":[36],"The":[37],"primary":[38],"challenges":[39],"include":[43],"memory":[44,84],"usage":[45],"and":[46,78,89,97,111],"speed.":[48],"This":[49],"paper":[50],"introduces":[51],"MNN-LLM,":[52],"framework":[54],"specifically":[55],"designed":[56],"accelerate":[58],"the":[59,70],"deployment":[60],"large":[62],"on":[65,92],"mobile":[66,93],"devices.":[67],"MNN-LLM":[68,118],"addresses":[69],"runtime":[71],"characteristics":[72,99],"LLMs":[74],"through":[75],"model":[76],"quantization":[77],"DRAM-Flash":[79],"hybrid":[80],"storage,":[81],"effectively":[82],"reducing":[83],"usage.":[85],"It":[86],"rearranges":[87],"weights":[88],"inputs":[90],"based":[91],"CPU":[94],"instruction":[95],"sets":[96],"GPU":[98],"while":[100],"employing":[101],"strategies":[102],"such":[103],"as":[104],"multicore":[105],"load":[106],"balancing,":[107],"mixed-precision":[108],"floating-point":[109],"operations,":[110],"geometric":[112],"computations":[113],"enhance":[115],"performance.":[116],"Notably,":[117],"achieves":[119],"up":[120],"8.6x":[123],"speed":[124],"increase":[125],"compared":[126],"current":[128],"mainstream":[129],"LLM-specific":[130],"frameworks.":[131]},"counts_by_year":[{"year":2026,"cited_by_count":5},{"year":2025,"cited_by_count":4}],"updated_date":"2026-06-13T07:54:00.901334","created_date":"2025-10-10T00:00:00"}
