{"id":"https://openalex.org/W7140234013","doi":"https://doi.org/10.48550/arxiv.2603.20492","title":"AE-LLM: Adaptive Efficiency Optimization for Large Language Models","display_name":"AE-LLM: Adaptive Efficiency Optimization for Large Language Models","publication_year":2026,"publication_date":"2026-03-20","ids":{"openalex":"https://openalex.org/W7140234013","doi":"https://doi.org/10.48550/arxiv.2603.20492"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2603.20492","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.20492","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2603.20492","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5113377241","display_name":"Kaito Tanaka","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Tanaka, Kaito","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5146793011","display_name":"Masato Ito","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ito, Masato","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5130468460","display_name":"Yuji Nishimura","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Nishimura, Yuji","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5130468734","display_name":"Keisuke Matsuda","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Matsuda, Keisuke","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5130437321","display_name":"Aya Nakayama","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Nakayama, Aya","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":null,"last_page":null},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.1754000037908554,"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":0.1754000037908554,"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"}},{"id":"https://openalex.org/T11636","display_name":"Artificial Intelligence in Healthcare and Education","score":0.125,"subfield":{"id":"https://openalex.org/subfields/2718","display_name":"Health Informatics"},"field":{"id":"https://openalex.org/fields/27","display_name":"Medicine"},"domain":{"id":"https://openalex.org/domains/4","display_name":"Health Sciences"}},{"id":"https://openalex.org/T10181","display_name":"Natural Language Processing Techniques","score":0.10279999673366547,"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/efficient-energy-use","display_name":"Efficient energy use","score":0.7253999710083008},{"id":"https://openalex.org/keywords/software-deployment","display_name":"Software deployment","score":0.6371999979019165},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.5612999796867371},{"id":"https://openalex.org/keywords/task","display_name":"Task (project management)","score":0.5479999780654907},{"id":"https://openalex.org/keywords/quantization","display_name":"Quantization (signal processing)","score":0.41690000891685486},{"id":"https://openalex.org/keywords/scheduling","display_name":"Scheduling (production processes)","score":0.3792000114917755},{"id":"https://openalex.org/keywords/resource-efficiency","display_name":"Resource efficiency","score":0.37610000371932983}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7657999992370605},{"id":"https://openalex.org/C2742236","wikidata":"https://www.wikidata.org/wiki/Q924713","display_name":"Efficient energy use","level":2,"score":0.7253999710083008},{"id":"https://openalex.org/C105339364","wikidata":"https://www.wikidata.org/wiki/Q2297740","display_name":"Software deployment","level":2,"score":0.6371999979019165},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.5612999796867371},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.5479999780654907},{"id":"https://openalex.org/C28855332","wikidata":"https://www.wikidata.org/wiki/Q198099","display_name":"Quantization (signal processing)","level":2,"score":0.41690000891685486},{"id":"https://openalex.org/C206729178","wikidata":"https://www.wikidata.org/wiki/Q2271896","display_name":"Scheduling (production processes)","level":2,"score":0.3792000114917755},{"id":"https://openalex.org/C2777958785","wikidata":"https://www.wikidata.org/wiki/Q17120940","display_name":"Resource efficiency","level":2,"score":0.37610000371932983},{"id":"https://openalex.org/C3018263672","wikidata":"https://www.wikidata.org/wiki/Q1296251","display_name":"Efficient algorithm","level":2,"score":0.3643999993801117},{"id":"https://openalex.org/C206345919","wikidata":"https://www.wikidata.org/wiki/Q20380951","display_name":"Resource (disambiguation)","level":2,"score":0.3538999855518341},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3447999954223633},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.32739999890327454},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.31679999828338623},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.3043999969959259},{"id":"https://openalex.org/C29202148","wikidata":"https://www.wikidata.org/wiki/Q287260","display_name":"Resource allocation","level":2,"score":0.3018999993801117},{"id":"https://openalex.org/C137836250","wikidata":"https://www.wikidata.org/wiki/Q984063","display_name":"Optimization problem","level":2,"score":0.29179999232292175},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.2718000113964081},{"id":"https://openalex.org/C186370098","wikidata":"https://www.wikidata.org/wiki/Q442787","display_name":"Energy (signal processing)","level":2,"score":0.2662999927997589},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.26440000534057617},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.25949999690055847},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.2517000138759613}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2603.20492","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.20492","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2603.20492","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.20492","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"},"sustainable_development_goals":[{"display_name":"Affordable and clean energy","score":0.8805265426635742,"id":"https://metadata.un.org/sdg/7"}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Large":[0],"Language":[1],"Models":[2],"(LLMs)":[3],"have":[4,29],"achieved":[5],"remarkable":[6],"success":[7],"across":[8,128,139],"diverse":[9,146],"applications,":[10],"yet":[11],"their":[12],"deployment":[13,87],"remains":[14],"challenging":[15],"due":[16],"to":[17,85,169,178],"substantial":[18],"computational":[19],"costs,":[20],"memory":[21,101],"requirements,":[22],"and":[23,53,63,79,103,111,131,144],"energy":[24,104],"consumption.":[25],"Recent":[26],"empirical":[27],"studies":[28],"demonstrated":[30],"that":[31,76,96,120,149],"no":[32],"single":[33],"efficiency":[34,82,126,158,171,183,201],"technique":[35],"is":[36],"universally":[37],"optimal;":[38],"instead,":[39],"the":[40,122,195],"effectiveness":[41],"of":[42,125,154,166,199],"methods":[43],"such":[44],"as":[45],"efficient":[46,117],"attention":[47],"mechanisms,":[48],"mixture-of-experts":[49],"(MoE),":[50],"parameter-efficient":[51],"fine-tuning,":[52,130],"quantization":[54],"varies":[55],"significantly":[56],"depending":[57],"on":[58],"task":[59,112],"characteristics,":[60],"resource":[61],"constraints,":[62],"model":[64],"scales.":[65],"Building":[66],"upon":[67],"these":[68],"insights,":[69],"we":[70],"propose":[71],"AE-LLM,":[72],"a":[73,92],"unified":[74],"framework":[75,95,175],"automatically":[77],"selects":[78],"combines":[80],"optimal":[81],"techniques":[83,127],"tailored":[84],"specific":[86],"scenarios.":[88],"Our":[89,185],"approach":[90],"introduces":[91],"multi-objective":[93],"optimization":[94],"jointly":[97],"considers":[98],"accuracy,":[99],"latency,":[100],"footprint,":[102],"consumption,":[105],"while":[106,160],"accounting":[107],"for":[108,193],"hardware":[109],"constraints":[110],"requirements.":[113],"We":[114],"develop":[115],"an":[116,152,190],"search":[118],"algorithm":[119],"explores":[121],"combinatorial":[123],"space":[124],"architecture,":[129],"inference":[132],"stages,":[133],"identifying":[134],"Pareto-optimal":[135],"configurations.":[136,172],"Extensive":[137],"experiments":[138],"15":[140],"models":[141],"(0.5B-70B":[142],"parameters)":[143],"10":[145],"tasks":[147],"demonstrate":[148],"AE-LLM":[150],"achieves":[151],"average":[153],"$2.8\\times$":[155],"improvement":[156],"in":[157],"metrics":[159],"maintaining":[161],"competitive":[162],"accuracy":[163],"(within":[164],"1.2\\%":[165],"baseline),":[167],"compared":[168],"static":[170],"Furthermore,":[173],"our":[174],"generalizes":[176],"effectively":[177],"vision-language":[179],"models,":[180],"achieving":[181],"similar":[182],"gains.":[184],"contributions":[186],"provide":[187],"practitioners":[188],"with":[189],"automated":[191],"tool":[192],"navigating":[194],"complex":[196],"trade-off":[197],"landscape":[198],"LLM":[200],"optimization.":[202]},"counts_by_year":[],"updated_date":"2026-08-12T21:12:35.861297","created_date":"2026-03-25T00:00:00"}