{"id":"https://openalex.org/W4403006781","doi":"https://doi.org/10.1145/3676536.3676741","title":"AdapMoE: Adaptive Sensitivity-based Expert Gating and Management for Efficient MoE Inference","display_name":"AdapMoE: Adaptive Sensitivity-based Expert Gating and Management for Efficient MoE Inference","publication_year":2024,"publication_date":"2024-10-27","ids":{"openalex":"https://openalex.org/W4403006781","doi":"https://doi.org/10.1145/3676536.3676741"},"language":"en","primary_location":{"id":"doi:10.1145/3676536.3676741","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3676536.3676741","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3676536.3676741","source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 43rd IEEE/ACM International Conference on Computer-Aided Design","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["arxiv","crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://dl.acm.org/doi/pdf/10.1145/3676536.3676741","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5081222727","display_name":"Shuzhang Zhong","orcid":null},"institutions":[{"id":"https://openalex.org/I20231570","display_name":"Peking University","ror":"https://ror.org/02v51f717","country_code":"CN","type":"education","lineage":["https://openalex.org/I20231570"]},{"id":"https://openalex.org/I4210100255","display_name":"Beijing Academy of Artificial Intelligence","ror":"https://ror.org/016a74861","country_code":"CN","type":"other","lineage":["https://openalex.org/I4210100255"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Shuzhang Zhong","raw_affiliation_strings":["Institute for Artificial Intelligence, Peking University, Beijing, China","School of Integrated Circuit, Peking University, Beijing, China","Institute for Artificial Intelligence &"],"raw_orcid":"https://orcid.org/0009-0006-5478-3604","affiliations":[{"raw_affiliation_string":"Institute for Artificial Intelligence, Peking University, Beijing, China","institution_ids":["https://openalex.org/I4210100255","https://openalex.org/I20231570"]},{"raw_affiliation_string":"School of Integrated Circuit, Peking University, Beijing, China","institution_ids":["https://openalex.org/I20231570"]},{"raw_affiliation_string":"Institute for Artificial Intelligence &","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5064892269","display_name":"Ling Liang","orcid":"https://orcid.org/0000-0002-8534-6494"},"institutions":[{"id":"https://openalex.org/I20231570","display_name":"Peking University","ror":"https://ror.org/02v51f717","country_code":"CN","type":"education","lineage":["https://openalex.org/I20231570"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Ling Liang","raw_affiliation_strings":["School of Integrated Circuit, Peking University, Beijing, China","School of Integrated Circuits, Peking University, Beijing, China"],"raw_orcid":"https://orcid.org/0000-0002-8534-6494","affiliations":[{"raw_affiliation_string":"School of Integrated Circuit, Peking University, Beijing, China","institution_ids":["https://openalex.org/I20231570"]},{"raw_affiliation_string":"School of Integrated Circuits, Peking University, Beijing, China","institution_ids":["https://openalex.org/I20231570"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5002438548","display_name":"Yuan Wang","orcid":"https://orcid.org/0000-0002-4951-4286"},"institutions":[{"id":"https://openalex.org/I20231570","display_name":"Peking University","ror":"https://ror.org/02v51f717","country_code":"CN","type":"education","lineage":["https://openalex.org/I20231570"]},{"id":"https://openalex.org/I4210165198","display_name":"Beijing Advanced Sciences and Innovation Center","ror":"https://ror.org/05qm21180","country_code":"CN","type":"facility","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210165198"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yuan Wang","raw_affiliation_strings":["Beijing Advanced Innovation Center for Integrated Circuits, Beijing, China","School of Integrated Circuit, Peking University, Beijing, China","School of Integrated Circuits, Peking University, Beijing, China"],"raw_orcid":"https://orcid.org/0000-0002-4951-4286","affiliations":[{"raw_affiliation_string":"Beijing Advanced Innovation Center for Integrated Circuits, Beijing, China","institution_ids":["https://openalex.org/I4210165198"]},{"raw_affiliation_string":"School of Integrated Circuit, Peking University, Beijing, China","institution_ids":["https://openalex.org/I20231570"]},{"raw_affiliation_string":"School of Integrated Circuits, Peking University, Beijing, China","institution_ids":["https://openalex.org/I20231570"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5002760019","display_name":"Runsheng Wang","orcid":"https://orcid.org/0000-0002-7514-0767"},"institutions":[{"id":"https://openalex.org/I20231570","display_name":"Peking University","ror":"https://ror.org/02v51f717","country_code":"CN","type":"education","lineage":["https://openalex.org/I20231570"]},{"id":"https://openalex.org/I4210165198","display_name":"Beijing Advanced Sciences and Innovation Center","ror":"https://ror.org/05qm21180","country_code":"CN","type":"facility","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210165198"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Runsheng Wang","raw_affiliation_strings":["Beijing Advanced Innovation Center for Integrated Circuits, Beijing, China","Institute of Electronic Design Automation, Peking University, Wuxi, China","School of Integrated Circuit, Peking University, Beijing, China","School of Integrated Circuits, Peking University, Beijing, China"],"raw_orcid":"https://orcid.org/0000-0002-7514-0767","affiliations":[{"raw_affiliation_string":"Beijing Advanced Innovation Center for Integrated Circuits, Beijing, China","institution_ids":["https://openalex.org/I4210165198"]},{"raw_affiliation_string":"Institute of Electronic Design Automation, Peking University, Wuxi, China","institution_ids":["https://openalex.org/I20231570"]},{"raw_affiliation_string":"School of Integrated Circuit, Peking University, Beijing, China","institution_ids":["https://openalex.org/I20231570"]},{"raw_affiliation_string":"School of Integrated Circuits, Peking University, Beijing, China","institution_ids":["https://openalex.org/I20231570"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5062886480","display_name":"Ru Huang","orcid":"https://orcid.org/0000-0002-8146-4821"},"institutions":[{"id":"https://openalex.org/I20231570","display_name":"Peking University","ror":"https://ror.org/02v51f717","country_code":"CN","type":"education","lineage":["https://openalex.org/I20231570"]},{"id":"https://openalex.org/I4210165198","display_name":"Beijing Advanced Sciences and Innovation Center","ror":"https://ror.org/05qm21180","country_code":"CN","type":"facility","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210165198"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Ru Huang","raw_affiliation_strings":["Beijing Advanced Innovation Center for Integrated Circuits, Beijing, China","Institute of Electronic Design Automation, Peking University, Wuxi, China","School of Integrated Circuit, Peking University, Beijing, China","School of Integrated Circuits, Peking University, Beijing, China"],"raw_orcid":"https://orcid.org/0000-0002-8146-4821","affiliations":[{"raw_affiliation_string":"Beijing Advanced Innovation Center for Integrated Circuits, Beijing, China","institution_ids":["https://openalex.org/I4210165198"]},{"raw_affiliation_string":"Institute of Electronic Design Automation, Peking University, Wuxi, China","institution_ids":["https://openalex.org/I20231570"]},{"raw_affiliation_string":"School of Integrated Circuit, Peking University, Beijing, China","institution_ids":["https://openalex.org/I20231570"]},{"raw_affiliation_string":"School of Integrated Circuits, Peking University, Beijing, China","institution_ids":["https://openalex.org/I20231570"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5100457502","display_name":"Meng Li","orcid":"https://orcid.org/0000-0002-7212-2264"},"institutions":[{"id":"https://openalex.org/I20231570","display_name":"Peking University","ror":"https://ror.org/02v51f717","country_code":"CN","type":"education","lineage":["https://openalex.org/I20231570"]},{"id":"https://openalex.org/I4210100255","display_name":"Beijing Academy of Artificial Intelligence","ror":"https://ror.org/016a74861","country_code":"CN","type":"other","lineage":["https://openalex.org/I4210100255"]},{"id":"https://openalex.org/I4210165198","display_name":"Beijing Advanced Sciences and Innovation Center","ror":"https://ror.org/05qm21180","country_code":"CN","type":"facility","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210165198"]}],"countries":["CN"],"is_corresponding":true,"raw_author_name":"Meng Li","raw_affiliation_strings":["Beijing Advanced Innovation Center for Integrated Circuits, Beijing, China","Institute for Artificial Intelligence, Peking University, Beijing, China","School of Integrated Circuit, Peking University, Beijing, China","Institute for Artificial Intelligence &"],"raw_orcid":"https://orcid.org/0000-0002-7212-2264","affiliations":[{"raw_affiliation_string":"Beijing Advanced Innovation Center for Integrated Circuits, Beijing, China","institution_ids":["https://openalex.org/I4210165198"]},{"raw_affiliation_string":"Institute for Artificial Intelligence, Peking University, Beijing, China","institution_ids":["https://openalex.org/I4210100255","https://openalex.org/I20231570"]},{"raw_affiliation_string":"School of Integrated Circuit, Peking University, Beijing, China","institution_ids":["https://openalex.org/I20231570"]},{"raw_affiliation_string":"Institute for Artificial Intelligence &","institution_ids":[]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":3,"corresponding_author_ids":["https://openalex.org/A5100457502"],"corresponding_institution_ids":["https://openalex.org/I20231570","https://openalex.org/I4210100255","https://openalex.org/I4210165198"],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"cited_by_count":14,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"9"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10203","display_name":"Recommender Systems and Techniques","score":0.9980000257492065,"subfield":{"id":"https://openalex.org/subfields/1710","display_name":"Information Systems"},"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/T10203","display_name":"Recommender Systems and Techniques","score":0.9980000257492065,"subfield":{"id":"https://openalex.org/subfields/1710","display_name":"Information Systems"},"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/T10764","display_name":"Privacy-Preserving Technologies in Data","score":0.9965999722480774,"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/T10028","display_name":"Topic Modeling","score":0.986299991607666,"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/computer-science","display_name":"Computer science","score":0.8058782815933228},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.6983901262283325},{"id":"https://openalex.org/keywords/speedup","display_name":"Speedup","score":0.6916126608848572},{"id":"https://openalex.org/keywords/latency","display_name":"Latency (audio)","score":0.6112521290779114},{"id":"https://openalex.org/keywords/software-deployment","display_name":"Software deployment","score":0.5929638147354126},{"id":"https://openalex.org/keywords/cache","display_name":"Cache","score":0.5458593368530273},{"id":"https://openalex.org/keywords/sensitivity","display_name":"Sensitivity (control systems)","score":0.46835508942604065},{"id":"https://openalex.org/keywords/gating","display_name":"Gating","score":0.4124520421028137},{"id":"https://openalex.org/keywords/ibm","display_name":"IBM","score":0.4103025794029236},{"id":"https://openalex.org/keywords/software-engineering","display_name":"Software engineering","score":0.23911485075950623},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.21313750743865967},{"id":"https://openalex.org/keywords/computer-network","display_name":"Computer network","score":0.19451674818992615},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.1826448142528534},{"id":"https://openalex.org/keywords/engineering","display_name":"Engineering","score":0.10367944836616516}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8058782815933228},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.6983901262283325},{"id":"https://openalex.org/C68339613","wikidata":"https://www.wikidata.org/wiki/Q1549489","display_name":"Speedup","level":2,"score":0.6916126608848572},{"id":"https://openalex.org/C82876162","wikidata":"https://www.wikidata.org/wiki/Q17096504","display_name":"Latency (audio)","level":2,"score":0.6112521290779114},{"id":"https://openalex.org/C105339364","wikidata":"https://www.wikidata.org/wiki/Q2297740","display_name":"Software deployment","level":2,"score":0.5929638147354126},{"id":"https://openalex.org/C115537543","wikidata":"https://www.wikidata.org/wiki/Q165596","display_name":"Cache","level":2,"score":0.5458593368530273},{"id":"https://openalex.org/C21200559","wikidata":"https://www.wikidata.org/wiki/Q7451068","display_name":"Sensitivity (control systems)","level":2,"score":0.46835508942604065},{"id":"https://openalex.org/C194544171","wikidata":"https://www.wikidata.org/wiki/Q21105679","display_name":"Gating","level":2,"score":0.4124520421028137},{"id":"https://openalex.org/C70388272","wikidata":"https://www.wikidata.org/wiki/Q5968558","display_name":"IBM","level":2,"score":0.4103025794029236},{"id":"https://openalex.org/C115903868","wikidata":"https://www.wikidata.org/wiki/Q80993","display_name":"Software engineering","level":1,"score":0.23911485075950623},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.21313750743865967},{"id":"https://openalex.org/C31258907","wikidata":"https://www.wikidata.org/wiki/Q1301371","display_name":"Computer network","level":1,"score":0.19451674818992615},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.1826448142528534},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.10367944836616516},{"id":"https://openalex.org/C24326235","wikidata":"https://www.wikidata.org/wiki/Q126095","display_name":"Electronic engineering","level":1,"score":0.0},{"id":"https://openalex.org/C171250308","wikidata":"https://www.wikidata.org/wiki/Q11468","display_name":"Nanotechnology","level":1,"score":0.0},{"id":"https://openalex.org/C86803240","wikidata":"https://www.wikidata.org/wiki/Q420","display_name":"Biology","level":0,"score":0.0},{"id":"https://openalex.org/C76155785","wikidata":"https://www.wikidata.org/wiki/Q418","display_name":"Telecommunications","level":1,"score":0.0},{"id":"https://openalex.org/C192562407","wikidata":"https://www.wikidata.org/wiki/Q228736","display_name":"Materials science","level":0,"score":0.0},{"id":"https://openalex.org/C42407357","wikidata":"https://www.wikidata.org/wiki/Q521","display_name":"Physiology","level":1,"score":0.0}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1145/3676536.3676741","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3676536.3676741","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3676536.3676741","source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 43rd IEEE/ACM International Conference on Computer-Aided Design","raw_type":"proceedings-article"},{"id":"pmh:oai:arXiv.org:2408.10284","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2408.10284","pdf_url":"https://arxiv.org/pdf/2408.10284","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"}],"best_oa_location":{"id":"doi:10.1145/3676536.3676741","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3676536.3676741","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3676536.3676741","source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 43rd IEEE/ACM International Conference on Computer-Aided Design","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4403006781.pdf","grobid_xml":"https://content.openalex.org/works/W4403006781.grobid-xml"},"referenced_works_count":7,"referenced_works":["https://openalex.org/W2066334462","https://openalex.org/W2734941459","https://openalex.org/W3119866685","https://openalex.org/W3138582970","https://openalex.org/W3205803342","https://openalex.org/W4287391717","https://openalex.org/W4321636575"],"related_works":["https://openalex.org/W2058965144","https://openalex.org/W2164382479","https://openalex.org/W2146343568","https://openalex.org/W98480971","https://openalex.org/W2150291671","https://openalex.org/W2013643406","https://openalex.org/W2027972911","https://openalex.org/W2157978810","https://openalex.org/W4391547476","https://openalex.org/W2597809628"],"abstract_inverted_index":{"Mixture-of-Experts":[0],"(MoE)":[1],"models":[2,12],"are":[3],"designed":[4],"to":[5,31,60,85,103],"enhance":[6],"the":[7,17,62,68,87,106,123],"efficiency":[8],"of":[9,70,89,126],"large":[10],"language":[11],"(LLMs)":[13],"without":[14,136],"proportionally":[15],"increasing":[16],"computational":[18],"demands.":[19],"However,":[20],"their":[21],"deployment":[22],"on":[23,78,112],"edge":[24],"devices":[25],"still":[26],"faces":[27],"significant":[28],"challenges":[29],"due":[30],"high":[32],"on-demand":[33,63],"loading":[34,64,72,107],"overheads":[35],"from":[36],"managing":[37],"sparsely":[38],"activated":[39,90,127],"experts.":[40],"This":[41],"paper":[42],"introduces":[43],"AdapMoE,":[44],"an":[45],"algorithm-system":[46],"co-design":[47],"framework":[48],"for":[49],"efficient":[50],"MoE":[51],"inference.":[52],"AdapMoE":[53,117],"features":[54],"adaptive":[55],"expert":[56],"gating":[57],"and":[58,75,99,131],"management":[59,101],"reduce":[61,105],"overheads.":[65],"We":[66],"observe":[67],"heterogeneity":[69],"experts":[71,91,128],"across":[73],"layers":[74],"tokens,":[76],"based":[77],"which":[79],"we":[80,94,115],"propose":[81],"a":[82,133],"sensitivity-based":[83],"strategy":[84],"adjust":[86],"number":[88,125],"dynamically.":[92],"Meanwhile,":[93],"also":[95],"integrate":[96],"advanced":[97],"prefetching":[98],"cache":[100],"techniques":[102],"further":[104],"latency.":[108],"Through":[109],"comprehensive":[110],"evaluations":[111],"various":[113],"platforms,":[114],"demonstrate":[116],"consistently":[118],"outperforms":[119],"existing":[120],"techniques,":[121],"reducing":[122],"average":[124],"by":[129],"25%":[130],"achieving":[132],"1.35\u00d7":[134],"speedup":[135],"accuracy":[137],"degradation.":[138],"Code":[139],"is":[140],"available":[141],"at:":[142],"https://github.com/PKU-SEC-Lab/AdapMoE.":[143]},"counts_by_year":[{"year":2026,"cited_by_count":8},{"year":2025,"cited_by_count":6}],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2024-10-01T00:00:00"}
