{"id":"https://openalex.org/W4386260498","doi":"https://doi.org/10.1109/infocom53939.2023.10228874","title":"PipeMoE: Accelerating Mixture-of-Experts through Adaptive Pipelining","display_name":"PipeMoE: Accelerating Mixture-of-Experts through Adaptive Pipelining","publication_year":2023,"publication_date":"2023-05-17","ids":{"openalex":"https://openalex.org/W4386260498","doi":"https://doi.org/10.1109/infocom53939.2023.10228874"},"language":"en","primary_location":{"id":"doi:10.1109/infocom53939.2023.10228874","is_oa":false,"landing_page_url":"https://doi.org/10.1109/infocom53939.2023.10228874","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE INFOCOM 2023 - IEEE Conference on Computer Communications","raw_type":"proceedings-article"},"type":"article","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/A5016836702","display_name":"Shaohuai Shi","orcid":"https://orcid.org/0000-0002-1418-5160"},"institutions":[{"id":"https://openalex.org/I204983213","display_name":"Harbin Institute of Technology","ror":"https://ror.org/01yqg2h08","country_code":"CN","type":"education","lineage":["https://openalex.org/I204983213"]}],"countries":["CN"],"is_corresponding":true,"raw_author_name":"Shaohuai Shi","raw_affiliation_strings":["Harbin Institute of Technology,School of Computer Science and Technology,Shenzhen"],"affiliations":[{"raw_affiliation_string":"Harbin Institute of Technology,School of Computer Science and Technology,Shenzhen","institution_ids":["https://openalex.org/I204983213"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101238767","display_name":"Xinglin Pan","orcid":null},"institutions":[{"id":"https://openalex.org/I141568987","display_name":"Hong Kong Baptist University","ror":"https://ror.org/0145fw131","country_code":"HK","type":"education","lineage":["https://openalex.org/I141568987"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Xinglin Pan","raw_affiliation_strings":["Hong Kong Baptist University,Department of Computer Science"],"affiliations":[{"raw_affiliation_string":"Hong Kong Baptist University,Department of Computer Science","institution_ids":["https://openalex.org/I141568987"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100730785","display_name":"Xiaowen Chu","orcid":"https://orcid.org/0000-0001-9745-4372"},"institutions":[{"id":"https://openalex.org/I141568987","display_name":"Hong Kong Baptist University","ror":"https://ror.org/0145fw131","country_code":"HK","type":"education","lineage":["https://openalex.org/I141568987"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Xiaowen Chu","raw_affiliation_strings":["Hong Kong Baptist University,Department of Computer Science"],"affiliations":[{"raw_affiliation_string":"Hong Kong Baptist University,Department of Computer Science","institution_ids":["https://openalex.org/I141568987"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5100374506","display_name":"Bo Li","orcid":"https://orcid.org/0000-0003-2083-9105"},"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":"Bo Li","raw_affiliation_strings":["The Hong Kong University of Science and Technology,Department of Computer Science and Engineering"],"affiliations":[{"raw_affiliation_string":"The Hong Kong University of Science and Technology,Department of Computer Science and Engineering","institution_ids":["https://openalex.org/I200769079"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":4,"corresponding_author_ids":["https://openalex.org/A5016836702"],"corresponding_institution_ids":["https://openalex.org/I204983213"],"apc_list":null,"apc_paid":null,"fwci":3.8001,"has_fulltext":false,"cited_by_count":15,"citation_normalized_percentile":{"value":0.93514728,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":94,"max":99},"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"10"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T12101","display_name":"Advanced Bandit Algorithms Research","score":0.9983999729156494,"subfield":{"id":"https://openalex.org/subfields/1803","display_name":"Management Science and Operations Research"},"field":{"id":"https://openalex.org/fields/18","display_name":"Decision Sciences"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}},"topics":[{"id":"https://openalex.org/T12101","display_name":"Advanced Bandit Algorithms Research","score":0.9983999729156494,"subfield":{"id":"https://openalex.org/subfields/1803","display_name":"Management Science and Operations Research"},"field":{"id":"https://openalex.org/fields/18","display_name":"Decision Sciences"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}},{"id":"https://openalex.org/T12761","display_name":"Data Stream Mining Techniques","score":0.9958000183105469,"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/T11704","display_name":"Mobile Crowdsensing and Crowdsourcing","score":0.9948999881744385,"subfield":{"id":"https://openalex.org/subfields/1706","display_name":"Computer Science Applications"},"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":0.7303428649902344},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.6135560870170593},{"id":"https://openalex.org/keywords/software-pipelining","display_name":"Software pipelining","score":0.5807421803474426},{"id":"https://openalex.org/keywords/acceleration","display_name":"Acceleration","score":0.45022907853126526},{"id":"https://openalex.org/keywords/computer-architecture","display_name":"Computer architecture","score":0.3544313311576843},{"id":"https://openalex.org/keywords/programming-language","display_name":"Programming language","score":0.294111430644989},{"id":"https://openalex.org/keywords/software","display_name":"Software","score":0.09265735745429993}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7303428649902344},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.6135560870170593},{"id":"https://openalex.org/C188854837","wikidata":"https://www.wikidata.org/wiki/Q268469","display_name":"Software pipelining","level":3,"score":0.5807421803474426},{"id":"https://openalex.org/C117896860","wikidata":"https://www.wikidata.org/wiki/Q11376","display_name":"Acceleration","level":2,"score":0.45022907853126526},{"id":"https://openalex.org/C118524514","wikidata":"https://www.wikidata.org/wiki/Q173212","display_name":"Computer architecture","level":1,"score":0.3544313311576843},{"id":"https://openalex.org/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.294111430644989},{"id":"https://openalex.org/C2777904410","wikidata":"https://www.wikidata.org/wiki/Q7397","display_name":"Software","level":2,"score":0.09265735745429993},{"id":"https://openalex.org/C74650414","wikidata":"https://www.wikidata.org/wiki/Q11397","display_name":"Classical mechanics","level":1,"score":0.0},{"id":"https://openalex.org/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","level":0,"score":0.0}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1109/infocom53939.2023.10228874","is_oa":false,"landing_page_url":"https://doi.org/10.1109/infocom53939.2023.10228874","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE INFOCOM 2023 - IEEE Conference on Computer Communications","raw_type":"proceedings-article"},{"id":"pmh:oai:repository.hkust.edu.hk:1783.1-127301","is_oa":false,"landing_page_url":"https://repository.hkust.edu.hk/ir/Record/1783.1-127301","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":"","raw_type":"Conference paper"}],"best_oa_location":null,"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/7","score":0.6100000143051147,"display_name":"Affordable and clean energy"}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":62,"referenced_works":["https://openalex.org/W1813030236","https://openalex.org/W1975604800","https://openalex.org/W2133557963","https://openalex.org/W2150884987","https://openalex.org/W2166221608","https://openalex.org/W2168231600","https://openalex.org/W2884711234","https://openalex.org/W2933138175","https://openalex.org/W2962758826","https://openalex.org/W2963786636","https://openalex.org/W2965373594","https://openalex.org/W2975712713","https://openalex.org/W2995435108","https://openalex.org/W3040573126","https://openalex.org/W3047537431","https://openalex.org/W3147874613","https://openalex.org/W3193250980","https://openalex.org/W3205972749","https://openalex.org/W4220818654","https://openalex.org/W4220838824","https://openalex.org/W4220967350","https://openalex.org/W4221150592","https://openalex.org/W4224308101","https://openalex.org/W4224441437","https://openalex.org/W4226079124","https://openalex.org/W4226515448","https://openalex.org/W4237651079","https://openalex.org/W4281399297","https://openalex.org/W4281688600","https://openalex.org/W4281922990","https://openalex.org/W4287121196","https://openalex.org/W4287391717","https://openalex.org/W4292779060","https://openalex.org/W4293718192","https://openalex.org/W4293775970","https://openalex.org/W4295312788","https://openalex.org/W4298422451","https://openalex.org/W4301239768","https://openalex.org/W4321636575","https://openalex.org/W4385245566","https://openalex.org/W6684859321","https://openalex.org/W6732520560","https://openalex.org/W6739693220","https://openalex.org/W6739901393","https://openalex.org/W6748645090","https://openalex.org/W6753209298","https://openalex.org/W6762287338","https://openalex.org/W6766673545","https://openalex.org/W6766978945","https://openalex.org/W6778883912","https://openalex.org/W6780805062","https://openalex.org/W6788811087","https://openalex.org/W6793102544","https://openalex.org/W6796487566","https://openalex.org/W6802290083","https://openalex.org/W6810081322","https://openalex.org/W6810737565","https://openalex.org/W6810764171","https://openalex.org/W6811726652","https://openalex.org/W6838284419","https://openalex.org/W6839092827","https://openalex.org/W6839827798"],"related_works":["https://openalex.org/W3132682100","https://openalex.org/W2068350864","https://openalex.org/W2117014006","https://openalex.org/W4233815414","https://openalex.org/W2372170743","https://openalex.org/W2389666628","https://openalex.org/W1602521801","https://openalex.org/W2357498671","https://openalex.org/W2993981457","https://openalex.org/W2108168782"],"abstract_inverted_index":{"Large":[0],"models":[1,133,171],"have":[2],"attracted":[3],"much":[4],"attention":[5],"in":[6,66,82,108,139,197],"the":[7,17,55,78,87,94,122,148,153,185],"AI":[8],"area.":[9],"The":[10],"sparsely":[11],"activated":[12],"mixture-of-experts":[13],"(MoE)":[14],"technique":[15],"pushes":[16],"model":[18],"size":[19],"to":[20,47,75,84,120,146],"a":[21,24,98,173],"trillion-level":[22],"with":[23,36,162],"sub-linear":[25],"increase":[26],"of":[27,57,61],"computations":[28,81],"as":[29],"an":[30,115,143],"MoE":[31,62,83,110,140,165,192],"layer":[32],"can":[33],"be":[34,48],"equipped":[35],"many":[37],"separate":[38],"experts,":[39],"but":[40],"only":[41],"one":[42],"or":[43],"two":[44,168],"experts":[45,60],"need":[46],"trained":[49],"for":[50,134],"each":[51],"input":[52],"data.":[53],"However,":[54],"feature":[56],"dynamically":[58],"activating":[59],"introduces":[63],"extensive":[64,160],"communications":[65,79],"distributed":[67],"training.":[68],"In":[69],"this":[70,128],"work,":[71],"we":[72,91,113,130],"propose":[73],"PipeMoE":[74,181],"adaptively":[76],"pipeline":[77,100,149,187],"and":[80,136,141,167,189],"maximally":[85],"hide":[86],"communication":[88,137],"time.":[89,125,199],"Specifically,":[90],"first":[92],"identify":[93],"root":[95],"reason":[96],"why":[97],"higher":[99],"degree":[101,150,188],"does":[102],"not":[103],"always":[104,183],"achieve":[105],"better":[106],"performance":[107,132],"training":[109,123,193,198],"models.":[111],"Then":[112],"formulate":[114],"optimization":[116],"problem":[117],"that":[118,152,179],"aims":[119],"minimize":[121],"iteration":[124,154],"To":[126],"solve":[127],"problem,":[129],"build":[131],"computation":[135],"tasks":[138],"develop":[142],"optimal":[144],"solution":[145],"determine":[147],"such":[151],"time":[155],"is":[156],"minimal.":[157],"We":[158],"conduct":[159],"experiments":[161],"174":[163],"typical":[164],"layers":[166],"real-world":[169],"NLP":[170],"on":[172],"64-GPU":[174],"cluster.":[175],"Experimental":[176],"results":[177],"show":[178],"our":[180],"almost":[182],"chooses":[184],"best":[186],"outperforms":[190],"state-of-the-art":[191],"systems":[194],"by":[195],"5%-77%":[196],"\u00a9":[200],"2023":[201],"IEEE.":[202]},"counts_by_year":[{"year":2026,"cited_by_count":2},{"year":2025,"cited_by_count":11},{"year":2024,"cited_by_count":2}],"updated_date":"2026-03-25T14:56:36.534964","created_date":"2025-10-10T00:00:00"}