{"id":"https://openalex.org/W4401508569","doi":"https://doi.org/10.1109/infocom52122.2024.10621087","title":"OTAS: An Elastic Transformer Serving System via Token Adaptation","display_name":"OTAS: An Elastic Transformer Serving System via Token Adaptation","publication_year":2024,"publication_date":"2024-05-20","ids":{"openalex":"https://openalex.org/W4401508569","doi":"https://doi.org/10.1109/infocom52122.2024.10621087"},"language":"en","primary_location":{"id":"doi:10.1109/infocom52122.2024.10621087","is_oa":false,"landing_page_url":"https://doi.org/10.1109/infocom52122.2024.10621087","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE INFOCOM 2024 - IEEE Conference on Computer Communications","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/A5100730366","display_name":"Jinyu Chen","orcid":"https://orcid.org/0000-0003-2811-3958"},"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":"Jinyu Chen","raw_affiliation_strings":["The Hong Kong Polytechnic University,Department of Computing"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"The Hong Kong Polytechnic University,Department of Computing","institution_ids":["https://openalex.org/I14243506"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101830672","display_name":"Wenchao Xu","orcid":"https://orcid.org/0000-0002-3256-8261"},"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":"Wenchao Xu","raw_affiliation_strings":["The Hong Kong Polytechnic University,Department of Computing"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"The Hong Kong Polytechnic University,Department of Computing","institution_ids":["https://openalex.org/I14243506"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5075795671","display_name":"Zicong Hong","orcid":"https://orcid.org/0000-0001-5689-382X"},"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":"Zicong Hong","raw_affiliation_strings":["The Hong Kong Polytechnic University,Department of Computing"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"The Hong Kong Polytechnic University,Department of Computing","institution_ids":["https://openalex.org/I14243506"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5043464306","display_name":"Song Guo","orcid":"https://orcid.org/0000-0001-9831-2202"},"institutions":[{"id":"https://openalex.org/I200769079","display_name":"Hong Kong University of Science and Technology","ror":"https://ror.org/00q4vv597","country_code":"HK","type":"education","lineage":["https://openalex.org/I200769079"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Song Guo","raw_affiliation_strings":["The Hong Kong University of Science and Technology,Department of Computer Science and Engineering"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"The Hong Kong University of Science and Technology,Department of Computer Science and Engineering","institution_ids":["https://openalex.org/I200769079"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5027119797","display_name":"Haozhao Wang","orcid":"https://orcid.org/0000-0002-7591-5315"},"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":"Haozhao Wang","raw_affiliation_strings":["Huazhong University of Science and Technology,School of Computer Science and Technology"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Huazhong University of Science and Technology,School of Computer Science and Technology","institution_ids":["https://openalex.org/I47720641"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100436639","display_name":"Jie Zhang","orcid":"https://orcid.org/0000-0001-7750-2197"},"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":"Jie Zhang","raw_affiliation_strings":["The Hong Kong Polytechnic University,Department of Computing"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"The Hong Kong Polytechnic University,Department of Computing","institution_ids":["https://openalex.org/I14243506"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5077093242","display_name":"Deze Zeng","orcid":"https://orcid.org/0000-0003-3276-1202"},"institutions":[{"id":"https://openalex.org/I3125743391","display_name":"China University of Geosciences (Beijing)","ror":"https://ror.org/04q6c7p66","country_code":"CN","type":"education","lineage":["https://openalex.org/I3125743391"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Deze Zeng","raw_affiliation_strings":["China University of Geosciences,School of Computer Science"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"China University of Geosciences,School of Computer Science","institution_ids":["https://openalex.org/I3125743391"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":4,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":3,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"1021","last_page":"1030"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.9957000017166138,"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":0.9957000017166138,"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/T14011","display_name":"Elevator Systems and Control","score":0.9945999979972839,"subfield":{"id":"https://openalex.org/subfields/2207","display_name":"Control and Systems 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/T12111","display_name":"Industrial Vision Systems and Defect Detection","score":0.9890999794006348,"subfield":{"id":"https://openalex.org/subfields/2209","display_name":"Industrial and Manufacturing 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/security-token","display_name":"Security token","score":0.7224985361099243},{"id":"https://openalex.org/keywords/transformer","display_name":"Transformer","score":0.6652187705039978},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.6254070401191711},{"id":"https://openalex.org/keywords/adaptation","display_name":"Adaptation (eye)","score":0.4454096257686615},{"id":"https://openalex.org/keywords/electrical-engineering","display_name":"Electrical engineering","score":0.37067490816116333},{"id":"https://openalex.org/keywords/electronic-engineering","display_name":"Electronic engineering","score":0.325295090675354},{"id":"https://openalex.org/keywords/computer-network","display_name":"Computer network","score":0.2571067810058594},{"id":"https://openalex.org/keywords/voltage","display_name":"Voltage","score":0.1961936354637146},{"id":"https://openalex.org/keywords/engineering","display_name":"Engineering","score":0.18086189031600952},{"id":"https://openalex.org/keywords/physics","display_name":"Physics","score":0.12779396772384644}],"concepts":[{"id":"https://openalex.org/C48145219","wikidata":"https://www.wikidata.org/wiki/Q1335365","display_name":"Security token","level":2,"score":0.7224985361099243},{"id":"https://openalex.org/C66322947","wikidata":"https://www.wikidata.org/wiki/Q11658","display_name":"Transformer","level":3,"score":0.6652187705039978},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6254070401191711},{"id":"https://openalex.org/C139807058","wikidata":"https://www.wikidata.org/wiki/Q352374","display_name":"Adaptation (eye)","level":2,"score":0.4454096257686615},{"id":"https://openalex.org/C119599485","wikidata":"https://www.wikidata.org/wiki/Q43035","display_name":"Electrical engineering","level":1,"score":0.37067490816116333},{"id":"https://openalex.org/C24326235","wikidata":"https://www.wikidata.org/wiki/Q126095","display_name":"Electronic engineering","level":1,"score":0.325295090675354},{"id":"https://openalex.org/C31258907","wikidata":"https://www.wikidata.org/wiki/Q1301371","display_name":"Computer network","level":1,"score":0.2571067810058594},{"id":"https://openalex.org/C165801399","wikidata":"https://www.wikidata.org/wiki/Q25428","display_name":"Voltage","level":2,"score":0.1961936354637146},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.18086189031600952},{"id":"https://openalex.org/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","level":0,"score":0.12779396772384644},{"id":"https://openalex.org/C120665830","wikidata":"https://www.wikidata.org/wiki/Q14620","display_name":"Optics","level":1,"score":0.0}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1109/infocom52122.2024.10621087","is_oa":false,"landing_page_url":"https://doi.org/10.1109/infocom52122.2024.10621087","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE INFOCOM 2024 - IEEE Conference on Computer Communications","raw_type":"proceedings-article"},{"id":"pmh:oai:repository.hkust.edu.hk:1783.1-142021","is_oa":false,"landing_page_url":"http://repository.hkust.edu.hk/ir/Record/1783.1-142021","pdf_url":null,"source":{"id":"https://openalex.org/S4306401796","display_name":"Rare & Special e-Zone (The Hong Kong University of Science and Technology)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I200769079","host_organization_name":"Hong Kong University of Science and Technology","host_organization_lineage":["https://openalex.org/I200769079"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Conference paper"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[{"id":"https://openalex.org/F4320307285","display_name":"Impact Fund","ror":"https://ror.org/00jb20j87"},{"id":"https://openalex.org/F4320337504","display_name":"Research and Development","ror":"https://ror.org/027s68j25"}],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":40,"referenced_works":["https://openalex.org/W95608104","https://openalex.org/W2110171203","https://openalex.org/W2734941459","https://openalex.org/W2896457183","https://openalex.org/W2964108773","https://openalex.org/W2964194231","https://openalex.org/W3094502228","https://openalex.org/W3118608800","https://openalex.org/W3130689885","https://openalex.org/W3185341429","https://openalex.org/W3195577433","https://openalex.org/W3205898353","https://openalex.org/W3210129272","https://openalex.org/W4221146855","https://openalex.org/W4283206493","https://openalex.org/W4283218352","https://openalex.org/W4285197124","https://openalex.org/W4287203292","https://openalex.org/W4287391717","https://openalex.org/W4292779060","https://openalex.org/W4312651322","https://openalex.org/W4318541554","https://openalex.org/W4320060379","https://openalex.org/W4320722432","https://openalex.org/W4321853806","https://openalex.org/W4385245566","https://openalex.org/W4388874804","https://openalex.org/W6730956707","https://openalex.org/W6755207826","https://openalex.org/W6778883912","https://openalex.org/W6784333009","https://openalex.org/W6787972765","https://openalex.org/W6788811087","https://openalex.org/W6794559225","https://openalex.org/W6798686915","https://openalex.org/W6800751262","https://openalex.org/W6846577953","https://openalex.org/W6849514112","https://openalex.org/W6849928207","https://openalex.org/W7000926438"],"related_works":["https://openalex.org/W2748952813","https://openalex.org/W4388335561","https://openalex.org/W2970530566","https://openalex.org/W4288261899","https://openalex.org/W4307309205","https://openalex.org/W2967478618","https://openalex.org/W4385009901","https://openalex.org/W4385572700","https://openalex.org/W4304700937","https://openalex.org/W2864363823"],"abstract_inverted_index":{"Transformer":[0],"model":[1],"empowered":[2],"architectures":[3],"have":[4],"become":[5],"a":[6,38,52,93,151,181],"pillar":[7],"of":[8,37,156,183],"cloud":[9],"services":[10],"that":[11,99,190],"keeps":[12],"reshaping":[13],"our":[14],"society.":[15],"However,":[16],"the":[17,62,76,145,154,160,168,193],"dynamic":[18],"query":[19,117],"loads":[20,118],"and":[21,66,106,119,130,159,179],"heterogeneous":[22],"user":[23,121],"requirements":[24],"severely":[25],"challenge":[26],"current":[27],"transformer":[28,58,84],"serving":[29,79],"systems,":[30],"which":[31,188],"rely":[32],"on":[33],"pre-training":[34],"multiple":[35,186],"variants":[36],"foundation":[39],"model,":[40],"i.e.,":[41],"with":[42,115,126,139,185],"different":[43],"sizes,":[44],"to":[45,61,103,110,143,149],"accommodate":[46],"varying":[47],"service":[48],"demands.":[49],"Unfortunately,":[50],"such":[51],"mechanism":[53],"is":[54],"unsuitable":[55],"for":[56,83,162],"large":[57],"models":[59,85],"due":[60],"additional":[63],"training":[64],"costs":[65],"excessive":[67],"I/O":[68],"delay.":[69],"In":[70],"this":[71],"paper,":[72],"we":[73,123],"introduce":[74],"OTAS,":[75],"first":[77,135],"elastic":[78],"system":[80,194],"specially":[81],"tailored":[82],"by":[86,172,196],"exploring":[87],"lightweight":[88],"token":[89,97,132,157,169],"management.":[90],"We":[91,177],"develop":[92],"novel":[94],"idea":[95],"called":[96],"adaptation":[98],"adds":[100],"prompting":[101],"tokens":[102,109],"improve":[104,144],"accuracy":[105,163],"removes":[107],"redundant":[108],"accelerate":[111],"inference.":[112],"To":[113],"cope":[114],"fluctuating":[116],"diverse":[120],"requests,":[122],"enhance":[124],"OTAS":[125,134,165,184,191],"application-aware":[127],"selective":[128],"batching":[129],"online":[131],"adaptation.":[133],"batches":[136],"incoming":[137],"queries":[138],"similar":[140],"service-level":[141],"objectives":[142],"ingress":[146],"throughput.":[147],"Then,":[148],"strike":[150],"tradeoff":[152],"between":[153],"overhead":[155],"increment":[158],"potentials":[161],"improvement,":[164],"adaptively":[166],"adjusts":[167],"execution":[170],"strategy":[171],"solving":[173],"an":[174],"optimization":[175],"problem.":[176],"implement":[178],"evaluate":[180],"prototype":[182],"datasets,":[187],"show":[189],"improves":[192],"utility":[195],"at":[197],"least":[198],"18.2%.":[199]},"counts_by_year":[{"year":2026,"cited_by_count":1},{"year":2025,"cited_by_count":1},{"year":2024,"cited_by_count":1}],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-10-10T00:00:00"}