{"id":"https://openalex.org/W7166873955","doi":"https://doi.org/10.18653/v1/2026.findings-acl.1946","title":"H-MAS: Hierarchical Multi-Agent Scheduling for Multi-Tenant LLM Serving","display_name":"H-MAS: Hierarchical Multi-Agent Scheduling for Multi-Tenant LLM Serving","publication_year":2026,"publication_date":"2026-01-01","ids":{"openalex":"https://openalex.org/W7166873955","doi":"https://doi.org/10.18653/v1/2026.findings-acl.1946"},"language":null,"primary_location":{"id":"doi:10.18653/v1/2026.findings-acl.1946","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.findings-acl.1946","pdf_url":"https://aclanthology.org/2026.findings-acl.1946.pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Findings of the Association for Computational Linguistics: ACL 2026","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://aclanthology.org/2026.findings-acl.1946.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5139762268","display_name":"Yuhan Liu","orcid":null},"institutions":[{"id":"https://openalex.org/I25254941","display_name":"Beijing Normal University","ror":"https://ror.org/022k4wk35","country_code":"CN","type":"education","lineage":["https://openalex.org/I25254941"]},{"id":"https://openalex.org/I4210108204","display_name":"China Electronic Information Industry Development","ror":"https://ror.org/01mbbxx50","country_code":"CN","type":"other","lineage":["https://openalex.org/I4210108204","https://openalex.org/I890469752"]},{"id":"https://openalex.org/I4210158311","display_name":"Chemical Industry Press","ror":"https://ror.org/05901gg63","country_code":"CN","type":"other","lineage":["https://openalex.org/I4210158311"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yuhan Liu","raw_affiliation_strings":["Beijing Normal University","Inspur (Beijing ) Electronic Information Industry Company"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Beijing Normal University","institution_ids":["https://openalex.org/I25254941"]},{"raw_affiliation_string":"Inspur (Beijing ) Electronic Information Industry Company","institution_ids":["https://openalex.org/I4210108204","https://openalex.org/I4210158311"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139837675","display_name":"Cong Xu","orcid":null},"institutions":[{"id":"https://openalex.org/I4210108204","display_name":"China Electronic Information Industry Development","ror":"https://ror.org/01mbbxx50","country_code":"CN","type":"other","lineage":["https://openalex.org/I4210108204","https://openalex.org/I890469752"]},{"id":"https://openalex.org/I4210158311","display_name":"Chemical Industry Press","ror":"https://ror.org/05901gg63","country_code":"CN","type":"other","lineage":["https://openalex.org/I4210158311"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Cong Xu","raw_affiliation_strings":["Inspur (Beijing ) Electronic Information Industry Company"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Inspur (Beijing ) Electronic Information Industry Company","institution_ids":["https://openalex.org/I4210108204","https://openalex.org/I4210158311"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139773480","display_name":"Qi Jia","orcid":null},"institutions":[{"id":"https://openalex.org/I4210108204","display_name":"China Electronic Information Industry Development","ror":"https://ror.org/01mbbxx50","country_code":"CN","type":"other","lineage":["https://openalex.org/I4210108204","https://openalex.org/I890469752"]},{"id":"https://openalex.org/I4210158311","display_name":"Chemical Industry Press","ror":"https://ror.org/05901gg63","country_code":"CN","type":"other","lineage":["https://openalex.org/I4210158311"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Qi Jia","raw_affiliation_strings":["Inspur (Beijing ) Electronic Information Industry Company"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Inspur (Beijing ) Electronic Information Industry Company","institution_ids":["https://openalex.org/I4210108204","https://openalex.org/I4210158311"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139794957","display_name":"Yihua Wang","orcid":null},"institutions":[{"id":"https://openalex.org/I25254941","display_name":"Beijing Normal University","ror":"https://ror.org/022k4wk35","country_code":"CN","type":"education","lineage":["https://openalex.org/I25254941"]},{"id":"https://openalex.org/I4210108204","display_name":"China Electronic Information Industry Development","ror":"https://ror.org/01mbbxx50","country_code":"CN","type":"other","lineage":["https://openalex.org/I4210108204","https://openalex.org/I890469752"]},{"id":"https://openalex.org/I4210158311","display_name":"Chemical Industry Press","ror":"https://ror.org/05901gg63","country_code":"CN","type":"other","lineage":["https://openalex.org/I4210158311"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yihua Wang","raw_affiliation_strings":["Beijing Normal University","Inspur (Beijing ) Electronic Information Industry Company"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Beijing Normal University","institution_ids":["https://openalex.org/I25254941"]},{"raw_affiliation_string":"Inspur (Beijing ) Electronic Information Industry Company","institution_ids":["https://openalex.org/I4210108204","https://openalex.org/I4210158311"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139748280","display_name":"Feiyu Chen","orcid":null},"institutions":[{"id":"https://openalex.org/I25254941","display_name":"Beijing Normal University","ror":"https://ror.org/022k4wk35","country_code":"CN","type":"education","lineage":["https://openalex.org/I25254941"]},{"id":"https://openalex.org/I4210108204","display_name":"China Electronic Information Industry Development","ror":"https://ror.org/01mbbxx50","country_code":"CN","type":"other","lineage":["https://openalex.org/I4210108204","https://openalex.org/I890469752"]},{"id":"https://openalex.org/I4210158311","display_name":"Chemical Industry Press","ror":"https://ror.org/05901gg63","country_code":"CN","type":"other","lineage":["https://openalex.org/I4210158311"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Feiyu Chen","raw_affiliation_strings":["Beijing Normal University","Inspur (Beijing ) Electronic Information Industry Company"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Beijing Normal University","institution_ids":["https://openalex.org/I25254941"]},{"raw_affiliation_string":"Inspur (Beijing ) Electronic Information Industry Company","institution_ids":["https://openalex.org/I4210108204","https://openalex.org/I4210158311"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139843161","display_name":"Liang Jin","orcid":null},"institutions":[{"id":"https://openalex.org/I4210108204","display_name":"China Electronic Information Industry Development","ror":"https://ror.org/01mbbxx50","country_code":"CN","type":"other","lineage":["https://openalex.org/I4210108204","https://openalex.org/I890469752"]},{"id":"https://openalex.org/I4210158311","display_name":"Chemical Industry Press","ror":"https://ror.org/05901gg63","country_code":"CN","type":"other","lineage":["https://openalex.org/I4210158311"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Liang Jin","raw_affiliation_strings":["Inspur (Beijing ) Electronic Information Industry Company"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Inspur (Beijing ) Electronic Information Industry Company","institution_ids":["https://openalex.org/I4210108204","https://openalex.org/I4210158311"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139826911","display_name":"Lu Liu","orcid":null},"institutions":[{"id":"https://openalex.org/I4210108204","display_name":"China Electronic Information Industry Development","ror":"https://ror.org/01mbbxx50","country_code":"CN","type":"other","lineage":["https://openalex.org/I4210108204","https://openalex.org/I890469752"]},{"id":"https://openalex.org/I4210158311","display_name":"Chemical Industry Press","ror":"https://ror.org/05901gg63","country_code":"CN","type":"other","lineage":["https://openalex.org/I4210158311"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Lu Liu","raw_affiliation_strings":["Inspur (Beijing ) Electronic Information Industry Company"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Inspur (Beijing ) Electronic Information Industry Company","institution_ids":["https://openalex.org/I4210108204","https://openalex.org/I4210158311"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139747312","display_name":"Yaqian Zhao","orcid":null},"institutions":[{"id":"https://openalex.org/I4210108204","display_name":"China Electronic Information Industry Development","ror":"https://ror.org/01mbbxx50","country_code":"CN","type":"other","lineage":["https://openalex.org/I4210108204","https://openalex.org/I890469752"]},{"id":"https://openalex.org/I4210158311","display_name":"Chemical Industry Press","ror":"https://ror.org/05901gg63","country_code":"CN","type":"other","lineage":["https://openalex.org/I4210158311"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yaqian Zhao","raw_affiliation_strings":["Inspur (Beijing ) Electronic Information Industry Company"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Inspur (Beijing ) Electronic Information Industry Company","institution_ids":["https://openalex.org/I4210108204","https://openalex.org/I4210158311"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139831023","display_name":"Yuting Ding","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yuting Ding","raw_affiliation_strings":["Southeast University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Southeast University","institution_ids":[]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5139835804","display_name":"Xiang Lorraine Li","orcid":null},"institutions":[{"id":"https://openalex.org/I152269853","display_name":"Qilu University of Technology","ror":"https://ror.org/04hyzq608","country_code":"CN","type":"education","lineage":["https://openalex.org/I152269853"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xiang Li","raw_affiliation_strings":["Key Laboratory of Computing Power Network and Information Security , Ministry of Education , Qilu University of Technology (Shandong Academy of Sciences)"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Key Laboratory of Computing Power Network and Information Security , Ministry of Education , Qilu University of Technology (Shandong Academy of Sciences)","institution_ids":["https://openalex.org/I152269853"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":4,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":true,"cited_by_count":0,"citation_normalized_percentile":{"value":0.8278953,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"39051","last_page":"39071"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11125","display_name":"Petri Nets in System Modeling","score":0.13120000064373016,"subfield":{"id":"https://openalex.org/subfields/1703","display_name":"Computational Theory and Mathematics"},"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/T11125","display_name":"Petri Nets in System Modeling","score":0.13120000064373016,"subfield":{"id":"https://openalex.org/subfields/1703","display_name":"Computational Theory and Mathematics"},"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/T12203","display_name":"Mobile Agent-Based Network Management","score":0.07190000265836716,"subfield":{"id":"https://openalex.org/subfields/1705","display_name":"Computer Networks and Communications"},"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/T10142","display_name":"Formal Methods in Verification","score":0.05090000107884407,"subfield":{"id":"https://openalex.org/subfields/1703","display_name":"Computational Theory and Mathematics"},"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/scheduling","display_name":"Scheduling (production processes)","score":0.5853999853134155},{"id":"https://openalex.org/keywords/job-shop-scheduling","display_name":"Job shop scheduling","score":0.28299999237060547},{"id":"https://openalex.org/keywords/key","display_name":"Key (lock)","score":0.22849999368190765},{"id":"https://openalex.org/keywords/fair-share-scheduling","display_name":"Fair-share scheduling","score":0.21439999341964722}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6794999837875366},{"id":"https://openalex.org/C206729178","wikidata":"https://www.wikidata.org/wiki/Q2271896","display_name":"Scheduling (production processes)","level":2,"score":0.5853999853134155},{"id":"https://openalex.org/C79403827","wikidata":"https://www.wikidata.org/wiki/Q3988","display_name":"Real-time computing","level":1,"score":0.3709999918937683},{"id":"https://openalex.org/C31258907","wikidata":"https://www.wikidata.org/wiki/Q1301371","display_name":"Computer network","level":1,"score":0.3547999858856201},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.28999999165534973},{"id":"https://openalex.org/C55416958","wikidata":"https://www.wikidata.org/wiki/Q6206757","display_name":"Job shop scheduling","level":3,"score":0.28299999237060547},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.22849999368190765},{"id":"https://openalex.org/C31689143","wikidata":"https://www.wikidata.org/wiki/Q733809","display_name":"Fair-share scheduling","level":3,"score":0.21439999341964722},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.2053000032901764},{"id":"https://openalex.org/C98045186","wikidata":"https://www.wikidata.org/wiki/Q205663","display_name":"Process (computing)","level":2,"score":0.19030000269412994}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.18653/v1/2026.findings-acl.1946","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.findings-acl.1946","pdf_url":"https://aclanthology.org/2026.findings-acl.1946.pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Findings of the Association for Computational Linguistics: ACL 2026","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.18653/v1/2026.findings-acl.1946","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.findings-acl.1946","pdf_url":"https://aclanthology.org/2026.findings-acl.1946.pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Findings of the Association for Computational Linguistics: ACL 2026","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G3875014413","display_name":null,"funder_award_id":"ZR2023LZH012","funder_id":"https://openalex.org/F4320324174","funder_display_name":"Natural Science Foundation of Shandong Province"},{"id":"https://openalex.org/G8665074011","display_name":null,"funder_award_id":"2023PY001","funder_id":"https://openalex.org/F4320324174","funder_display_name":"Natural Science Foundation of Shandong Province"}],"funders":[{"id":"https://openalex.org/F4320324174","display_name":"Natural Science Foundation of Shandong Province","ror":null},{"id":"https://openalex.org/F4320324887","display_name":"Shandong Academy of Sciences","ror":"https://ror.org/04y8d6y55"},{"id":"https://openalex.org/F4320328720","display_name":"Qilu University of Technology","ror":"https://ror.org/04hyzq608"}],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7166873955.pdf","grobid_xml":"https://content.openalex.org/works/W7166873955.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Multi-tenant":[0],"Model-as-a-Service":[1],"(MaaS)":[2],"LLM":[3],"serving":[4],"must":[5],"maintain":[6],"stringent":[7],"quality":[8],"of":[9],"service":[10],"(QoS)":[11],"despite":[12],"heterogeneous":[13,158],"requests":[14],"competing":[15],"for":[16],"constrained":[17],"GPU":[18],"resources.In":[19],"practice,":[20],"MaaS":[21],"workloads":[22,38],"exhibit":[23],"non-stationarity":[24],"across":[25,40],"multiple":[26,41],"time":[27,42],"scales,":[28,43],"including":[29],"request":[30,45,86,155],"bursts,":[31],"request-composition":[32],"drift,":[33,153],"and":[34,88,114,121,133,144,146,157],"persistent":[35],"workload":[36,152],"shifts.Because":[37],"change":[39],"existing":[44],"schedulers":[46],"often":[47],"rely":[48],"on":[49],"a":[50,70,77,81,91,102],"single":[51],"fixed":[52],"policy":[53,110],"(e.g.,":[54],"First-Come-First-Served,":[55],"FCFS)":[56],"that":[57,74,137],"remains":[58],"unchanged":[59],"at":[60],"runtime,":[61],"which":[62],"can":[63],"lead":[64],"to":[65],"unstable":[66],"QoS.We":[67],"propose":[68],"H-MAS,":[69],"hierarchical":[71,103],"multi-agent":[72],"scheduler":[73],"operates":[75],"in":[76],"layered":[78],"closed":[79],"loop:":[80],"perception/prediction":[82],"layer":[83,93,105,124],"infers":[84],"lightweight":[85],"attributes":[87],"cost":[89],"signals;":[90],"feedback":[92],"summarizes":[94],"runtime":[95],"metrics":[96],"into":[97],"short-and":[98],"long-horizon":[99],"QoS":[100,150],"states;":[101],"control":[104],"updates":[106],"the":[107],"active":[108],"scheduling":[109],"over":[111,118],"longer":[112],"horizons":[113],"tunes":[115],"execution":[116,123],"parameters":[117],"shorter":[119],"horizons;":[120],"an":[122],"applies":[125],"these":[126],"decisions":[127],"during":[128],"inference.Experiments":[129],"with":[130],"load":[131],"scaling":[132],"Azure-trace":[134],"replays":[135],"show":[136],"H-MAS":[138],"achieves":[139],"1.2\u00d7-3.0\u00d7higher":[140],"Goodput":[141],"than":[142],"SGLang":[143],"vLLM,":[145],"maintains":[147],"more":[148],"stable":[149],"under":[151],"diverse":[154],"lengths":[156],"SLO":[159],"targets.":[160]},"counts_by_year":[],"updated_date":"2026-08-25T07:29:55.448023","created_date":"2026-07-02T00:00:00"}
