{"id":"https://openalex.org/W7162799881","doi":"https://doi.org/10.1145/3745756.3809208","title":"Act Before It\u2019s Too Late: Power-Efficient LLM Inference on Mobile Device","display_name":"Act Before It\u2019s Too Late: Power-Efficient LLM Inference on Mobile Device","publication_year":2026,"publication_date":"2026-05-29","ids":{"openalex":"https://openalex.org/W7162799881","doi":"https://doi.org/10.1145/3745756.3809208"},"language":null,"primary_location":{"id":"doi:10.1145/3745756.3809208","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3745756.3809208","pdf_url":null,"source":null,"license":"cc-by-nc-nd","license_id":"https://openalex.org/licenses/cc-by-nc-nd","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 24th Annual International Conference on Mobile Systems, Applications and Services","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://doi.org/10.1145/3745756.3809208","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5122861175","display_name":"Haolin Chu","orcid":null},"institutions":[{"id":"https://openalex.org/I139759216","display_name":"Beijing University of Posts and Telecommunications","ror":"https://ror.org/04w9fbh59","country_code":"CN","type":"education","lineage":["https://openalex.org/I139759216"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Haolin Chu","raw_affiliation_strings":["Beijing University of Posts and Telecommunications, Beijing, China"],"raw_orcid":"https://orcid.org/0009-0000-7184-3150","affiliations":[{"raw_affiliation_string":"Beijing University of Posts and Telecommunications, Beijing, China","institution_ids":["https://openalex.org/I139759216"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100739537","display_name":"Jinxiao Fan","orcid":"https://orcid.org/0000-0002-0119-1916"},"institutions":[{"id":"https://openalex.org/I139759216","display_name":"Beijing University of Posts and Telecommunications","ror":"https://ror.org/04w9fbh59","country_code":"CN","type":"education","lineage":["https://openalex.org/I139759216"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Jinxiao Fan","raw_affiliation_strings":["Beijing University of Posts and Telecommunications, Beijing, China"],"raw_orcid":"https://orcid.org/0000-0002-0119-1916","affiliations":[{"raw_affiliation_string":"Beijing University of Posts and Telecommunications, Beijing, China","institution_ids":["https://openalex.org/I139759216"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5137353943","display_name":"Jiabin Deng","orcid":"https://orcid.org/0009-0001-1869-742X"},"institutions":[{"id":"https://openalex.org/I139759216","display_name":"Beijing University of Posts and Telecommunications","ror":"https://ror.org/04w9fbh59","country_code":"CN","type":"education","lineage":["https://openalex.org/I139759216"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Jiabin Deng","raw_affiliation_strings":["Beijing University of Posts and Telecommunications, Beijing, China"],"raw_orcid":"https://orcid.org/0009-0001-1869-742X","affiliations":[{"raw_affiliation_string":"Beijing University of Posts and Telecommunications, Beijing, China","institution_ids":["https://openalex.org/I139759216"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5137338965","display_name":"Bensong Yu","orcid":"https://orcid.org/0009-0005-6949-3995"},"institutions":[{"id":"https://openalex.org/I139759216","display_name":"Beijing University of Posts and Telecommunications","ror":"https://ror.org/04w9fbh59","country_code":"CN","type":"education","lineage":["https://openalex.org/I139759216"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Bensong Yu","raw_affiliation_strings":["Beijing University of Posts and Telecommunications, Beijing, China"],"raw_orcid":"https://orcid.org/0009-0005-6949-3995","affiliations":[{"raw_affiliation_string":"Beijing University of Posts and Telecommunications, Beijing, China","institution_ids":["https://openalex.org/I139759216"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5040158607","display_name":"Liguang Xie","orcid":"https://orcid.org/0000-0002-3764-9888"},"institutions":[{"id":"https://openalex.org/I4210091175","display_name":"Ascent Technology (United States)","ror":"https://ror.org/00cpm4162","country_code":"US","type":"company","lineage":["https://openalex.org/I4210091175"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Liguang Xie","raw_affiliation_strings":["Bytedance technology company Limited, Seattle, USA"],"raw_orcid":"https://orcid.org/0000-0002-3764-9888","affiliations":[{"raw_affiliation_string":"Bytedance technology company Limited, Seattle, USA","institution_ids":["https://openalex.org/I4210091175"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5137347955","display_name":"Liang Liu","orcid":"https://orcid.org/0000-0002-5040-2468"},"institutions":[{"id":"https://openalex.org/I139759216","display_name":"Beijing University of Posts and Telecommunications","ror":"https://ror.org/04w9fbh59","country_code":"CN","type":"education","lineage":["https://openalex.org/I139759216"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Liang Liu","raw_affiliation_strings":["Beijing University of Posts and Telecommunications, Beijing, China"],"raw_orcid":"https://orcid.org/0000-0002-5040-2468","affiliations":[{"raw_affiliation_string":"Beijing University of Posts and Telecommunications, Beijing, China","institution_ids":["https://openalex.org/I139759216"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5137339870","display_name":"Huadong Ma","orcid":"https://orcid.org/0000-0002-7199-5047"},"institutions":[{"id":"https://openalex.org/I139759216","display_name":"Beijing University of Posts and Telecommunications","ror":"https://ror.org/04w9fbh59","country_code":"CN","type":"education","lineage":["https://openalex.org/I139759216"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Huadong Ma","raw_affiliation_strings":["Beijing University of Posts and Telecommunications, Beijing, China"],"raw_orcid":"https://orcid.org/0000-0002-7199-5047","affiliations":[{"raw_affiliation_string":"Beijing University of Posts and Telecommunications, Beijing, China","institution_ids":["https://openalex.org/I139759216"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5137329204","display_name":"Xiaolong Zheng","orcid":"https://orcid.org/0000-0001-7950-6773"},"institutions":[{"id":"https://openalex.org/I139759216","display_name":"Beijing University of Posts and Telecommunications","ror":"https://ror.org/04w9fbh59","country_code":"CN","type":"education","lineage":["https://openalex.org/I139759216"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xiaolong Zheng","raw_affiliation_strings":["Beijing University of Posts and Telecommunications, Beijing, China"],"raw_orcid":"https://orcid.org/0000-0001-7950-6773","affiliations":[{"raw_affiliation_string":"Beijing University of Posts and Telecommunications, Beijing, China","institution_ids":["https://openalex.org/I139759216"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":8,"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":0.87399565,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"338","last_page":"351"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.2842000126838684,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"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/T10036","display_name":"Advanced Neural Network Applications","score":0.2842000126838684,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"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/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.2563999891281128,"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/T12238","display_name":"Green IT and Sustainability","score":0.04600000008940697,"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/inference","display_name":"Inference","score":0.7407000064849854},{"id":"https://openalex.org/keywords/mobile-device","display_name":"Mobile device","score":0.7200999855995178},{"id":"https://openalex.org/keywords/pipeline","display_name":"Pipeline (software)","score":0.6608999967575073},{"id":"https://openalex.org/keywords/kernel","display_name":"Kernel (algebra)","score":0.5769000053405762},{"id":"https://openalex.org/keywords/security-token","display_name":"Security token","score":0.5188999772071838},{"id":"https://openalex.org/keywords/dram","display_name":"Dram","score":0.44029998779296875},{"id":"https://openalex.org/keywords/mobile-computing","display_name":"Mobile computing","score":0.42890000343322754},{"id":"https://openalex.org/keywords/resource","display_name":"Resource (disambiguation)","score":0.37279999256134033}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8087000250816345},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.7407000064849854},{"id":"https://openalex.org/C186967261","wikidata":"https://www.wikidata.org/wiki/Q5082128","display_name":"Mobile device","level":2,"score":0.7200999855995178},{"id":"https://openalex.org/C43521106","wikidata":"https://www.wikidata.org/wiki/Q2165493","display_name":"Pipeline (software)","level":2,"score":0.6608999967575073},{"id":"https://openalex.org/C74193536","wikidata":"https://www.wikidata.org/wiki/Q574844","display_name":"Kernel (algebra)","level":2,"score":0.5769000053405762},{"id":"https://openalex.org/C48145219","wikidata":"https://www.wikidata.org/wiki/Q1335365","display_name":"Security token","level":2,"score":0.5188999772071838},{"id":"https://openalex.org/C7366592","wikidata":"https://www.wikidata.org/wiki/Q1255620","display_name":"Dram","level":2,"score":0.44029998779296875},{"id":"https://openalex.org/C144543869","wikidata":"https://www.wikidata.org/wiki/Q2738570","display_name":"Mobile computing","level":2,"score":0.42890000343322754},{"id":"https://openalex.org/C149635348","wikidata":"https://www.wikidata.org/wiki/Q193040","display_name":"Embedded system","level":1,"score":0.3840999901294708},{"id":"https://openalex.org/C206345919","wikidata":"https://www.wikidata.org/wiki/Q20380951","display_name":"Resource (disambiguation)","level":2,"score":0.37279999256134033},{"id":"https://openalex.org/C123657996","wikidata":"https://www.wikidata.org/wiki/Q12271","display_name":"Architecture","level":2,"score":0.34360000491142273},{"id":"https://openalex.org/C107598950","wikidata":"https://www.wikidata.org/wiki/Q259864","display_name":"Microarchitecture","level":2,"score":0.299699991941452},{"id":"https://openalex.org/C38652104","wikidata":"https://www.wikidata.org/wiki/Q3510521","display_name":"Computer security","level":1,"score":0.29580000042915344},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.290800005197525},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.2851000130176544},{"id":"https://openalex.org/C78766204","wikidata":"https://www.wikidata.org/wiki/Q555032","display_name":"Multi-core processor","level":2,"score":0.28349998593330383},{"id":"https://openalex.org/C79403827","wikidata":"https://www.wikidata.org/wiki/Q3988","display_name":"Real-time computing","level":1,"score":0.2808000147342682},{"id":"https://openalex.org/C95491727","wikidata":"https://www.wikidata.org/wiki/Q992968","display_name":"Mobile telephony","level":3,"score":0.2752000093460083},{"id":"https://openalex.org/C31258907","wikidata":"https://www.wikidata.org/wiki/Q1301371","display_name":"Computer network","level":1,"score":0.2696000039577484},{"id":"https://openalex.org/C2164484","wikidata":"https://www.wikidata.org/wiki/Q5170150","display_name":"Core (optical fiber)","level":2,"score":0.2669999897480011},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.2630000114440918},{"id":"https://openalex.org/C2779191767","wikidata":"https://www.wikidata.org/wiki/Q6887075","display_name":"Mobile cloud computing","level":3,"score":0.2624000012874603},{"id":"https://openalex.org/C176649486","wikidata":"https://www.wikidata.org/wiki/Q2308807","display_name":"Memory management","level":3,"score":0.2565000057220459},{"id":"https://openalex.org/C206588197","wikidata":"https://www.wikidata.org/wiki/Q846574","display_name":"Reuse","level":2,"score":0.2540999948978424}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3745756.3809208","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3745756.3809208","pdf_url":null,"source":null,"license":"cc-by-nc-nd","license_id":"https://openalex.org/licenses/cc-by-nc-nd","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 24th Annual International Conference on Mobile Systems, Applications and Services","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.1145/3745756.3809208","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3745756.3809208","pdf_url":null,"source":null,"license":"cc-by-nc-nd","license_id":"https://openalex.org/licenses/cc-by-nc-nd","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 24th Annual International Conference on Mobile Systems, Applications and Services","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":18,"referenced_works":["https://openalex.org/W1981723834","https://openalex.org/W2056947227","https://openalex.org/W2062240598","https://openalex.org/W2080256219","https://openalex.org/W2125381034","https://openalex.org/W2604319603","https://openalex.org/W2762077810","https://openalex.org/W2790957308","https://openalex.org/W2982083293","https://openalex.org/W3125571220","https://openalex.org/W4242473187","https://openalex.org/W4253863518","https://openalex.org/W4387321091","https://openalex.org/W4395106452","https://openalex.org/W4399074329","https://openalex.org/W4404918643","https://openalex.org/W4405805825","https://openalex.org/W7123362947"],"related_works":[],"abstract_inverted_index":{"This":[0],"paper":[1],"presents":[2],"TurboInfer,":[3],"a":[4],"system":[5],"that":[6,20],"enables":[7],"power-efficient":[8],"LLM":[9,79],"inference":[10,33,80],"on":[11],"mobile":[12,60,75],"devices.":[13],"The":[14],"core":[15],"insight":[16],"behind":[17],"TurboInfer":[18],"is":[19],"while":[21],"LLMs":[22],"are":[23],"power-intensive":[24],"due":[25],"to":[26,59,69,83],"their":[27],"heavy":[28],"computational":[29],"demands,":[30],"the":[31,53,63,70,100],"model":[32],"experiences":[34],"unavoidable":[35],"GPU":[36,49,92],"stalls":[37,50,93],"caused":[38],"by":[39],"tensor":[40],"preparation":[41],"for":[42,96],"subsequent":[43],"kernel":[44,86],"executions":[45],"at":[46],"run-time.":[47],"These":[48],"arise":[51],"from":[52],"unique":[54],"host-controlled":[55],"execution":[56],"pipeline":[57],"tailored":[58],"phones":[61],"and":[62],"significant":[64],"DRAM":[65],"access":[66],"contention":[67],"inherent":[68],"shared":[71],"memory":[72],"architecture":[73],"of":[74,85,99],"System-on-chips":[76],"(SoCs).":[77],"With":[78],"requiring":[81],"hundreds":[82],"thousands":[84],"executions,":[87],"these":[88],"short":[89],"but":[90],"frequent":[91],"accumulate,":[94],"accounting":[95],"over":[97],"74%":[98],"token":[101],"generation":[102],"latency.":[103]},"counts_by_year":[],"updated_date":"2026-06-11T09:08:48.828518","created_date":"2026-05-30T00:00:00"}