{"id":"https://openalex.org/W4414198120","doi":"https://doi.org/10.1109/dac63849.2025.11132853","title":"BirdMoE: Reducing Communication Costs for Mixture-of-Experts Training Using Load-Aware Bi-random Quantization","display_name":"BirdMoE: Reducing Communication Costs for Mixture-of-Experts Training Using Load-Aware Bi-random Quantization","publication_year":2025,"publication_date":"2025-06-22","ids":{"openalex":"https://openalex.org/W4414198120","doi":"https://doi.org/10.1109/dac63849.2025.11132853"},"language":"en","primary_location":{"id":"doi:10.1109/dac63849.2025.11132853","is_oa":false,"landing_page_url":"https://doi.org/10.1109/dac63849.2025.11132853","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 62nd ACM/IEEE Design Automation Conference (DAC)","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/A5069261286","display_name":"Donglei Wu","orcid":"https://orcid.org/0000-0003-0358-0533"},"institutions":[{"id":"https://openalex.org/I37987034","display_name":"Guangzhou University","ror":"https://ror.org/05ar8rn06","country_code":"CN","type":"education","lineage":["https://openalex.org/I37987034"]}],"countries":["CN"],"is_corresponding":true,"raw_author_name":"Donglei Wu","raw_affiliation_strings":["Guangzhou University,Cyberspace Institute of Advanced Technology,Guangzhou,China"],"affiliations":[{"raw_affiliation_string":"Guangzhou University,Cyberspace Institute of Advanced Technology,Guangzhou,China","institution_ids":["https://openalex.org/I37987034"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101636404","display_name":"Weihao Yang","orcid":"https://orcid.org/0009-0002-6337-1768"},"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":false,"raw_author_name":"Weihao Yang","raw_affiliation_strings":["Harbin Institute of Technology,School of Computer Science and Technology,Shenzhen,China"],"affiliations":[{"raw_affiliation_string":"Harbin Institute of Technology,School of Computer Science and Technology,Shenzhen,China","institution_ids":["https://openalex.org/I204983213"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5021153181","display_name":"Xiangyu Zou","orcid":"https://orcid.org/0000-0001-5104-8301"},"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":false,"raw_author_name":"Xiangyu Zou","raw_affiliation_strings":["Harbin Institute of Technology,School of Computer Science and Technology,Shenzhen,China"],"affiliations":[{"raw_affiliation_string":"Harbin Institute of Technology,School of Computer Science and Technology,Shenzhen,China","institution_ids":["https://openalex.org/I204983213"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5109706936","display_name":"Jinda Jia","orcid":"https://orcid.org/0009-0008-5283-5241"},"institutions":[{"id":"https://openalex.org/I4210119109","display_name":"Indiana University Bloomington","ror":"https://ror.org/02k40bc56","country_code":"US","type":"education","lineage":["https://openalex.org/I4210119109","https://openalex.org/I592451"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Jinda Jia","raw_affiliation_strings":["Indiana University,Luddy School of Informatics, Computing, and Engineering,Bloomington,IN,USA"],"affiliations":[{"raw_affiliation_string":"Indiana University,Luddy School of Informatics, Computing, and Engineering,Bloomington,IN,USA","institution_ids":["https://openalex.org/I4210119109"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5063703614","display_name":"Dingwen Tao","orcid":"https://orcid.org/0000-0001-5422-4497"},"institutions":[{"id":"https://openalex.org/I19820366","display_name":"Chinese Academy of Sciences","ror":"https://ror.org/034t30j35","country_code":"CN","type":"funder","lineage":["https://openalex.org/I19820366"]},{"id":"https://openalex.org/I4210090176","display_name":"Institute of Computing Technology","ror":"https://ror.org/0090r4d87","country_code":"CN","type":"facility","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210090176"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Dingwen Tao","raw_affiliation_strings":["Institute of Computing Technology, Chinese Academy of Sciences,Beijing,China"],"affiliations":[{"raw_affiliation_string":"Institute of Computing Technology, Chinese Academy of Sciences,Beijing,China","institution_ids":["https://openalex.org/I4210090176","https://openalex.org/I19820366"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5050928208","display_name":"Wen Xia","orcid":"https://orcid.org/0000-0003-4093-6391"},"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":false,"raw_author_name":"Wen Xia","raw_affiliation_strings":["Harbin Institute of Technology,School of Computer Science and Technology,Shenzhen,China"],"affiliations":[{"raw_affiliation_string":"Harbin Institute of Technology,School of Computer Science and Technology,Shenzhen,China","institution_ids":["https://openalex.org/I204983213"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5056608045","display_name":"Zhihong Tian","orcid":"https://orcid.org/0000-0002-9409-5359"},"institutions":[{"id":"https://openalex.org/I37987034","display_name":"Guangzhou University","ror":"https://ror.org/05ar8rn06","country_code":"CN","type":"education","lineage":["https://openalex.org/I37987034"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zhihong Tian","raw_affiliation_strings":["Guangzhou University,Cyberspace Institute of Advanced Technology,Guangzhou,China"],"affiliations":[{"raw_affiliation_string":"Guangzhou University,Cyberspace Institute of Advanced Technology,Guangzhou,China","institution_ids":["https://openalex.org/I37987034"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":7,"corresponding_author_ids":["https://openalex.org/A5069261286"],"corresponding_institution_ids":["https://openalex.org/I37987034"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.23425839,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"7"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10326","display_name":"Indoor and Outdoor Localization Technologies","score":0.9710000157356262,"subfield":{"id":"https://openalex.org/subfields/2208","display_name":"Electrical and Electronic Engineering"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T10326","display_name":"Indoor and Outdoor Localization Technologies","score":0.9710000157356262,"subfield":{"id":"https://openalex.org/subfields/2208","display_name":"Electrical and Electronic 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/T12814","display_name":"Gaussian Processes and Bayesian Inference","score":0.9550999999046326,"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/T12879","display_name":"Distributed Sensor Networks and Detection Algorithms","score":0.9491999745368958,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/quantization","display_name":"Quantization (signal processing)","score":0.7506999969482422},{"id":"https://openalex.org/keywords/speedup","display_name":"Speedup","score":0.6541000008583069},{"id":"https://openalex.org/keywords/training","display_name":"Training (meteorology)","score":0.5633000135421753},{"id":"https://openalex.org/keywords/communications-system","display_name":"Communications system","score":0.42160001397132874},{"id":"https://openalex.org/keywords/data-compression","display_name":"Data compression","score":0.39259999990463257},{"id":"https://openalex.org/keywords/training-set","display_name":"Training set","score":0.3856000006198883},{"id":"https://openalex.org/keywords/software","display_name":"Software","score":0.3465000092983246}],"concepts":[{"id":"https://openalex.org/C28855332","wikidata":"https://www.wikidata.org/wiki/Q198099","display_name":"Quantization (signal processing)","level":2,"score":0.7506999969482422},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6948000192642212},{"id":"https://openalex.org/C68339613","wikidata":"https://www.wikidata.org/wiki/Q1549489","display_name":"Speedup","level":2,"score":0.6541000008583069},{"id":"https://openalex.org/C2777211547","wikidata":"https://www.wikidata.org/wiki/Q17141490","display_name":"Training (meteorology)","level":2,"score":0.5633000135421753},{"id":"https://openalex.org/C101765175","wikidata":"https://www.wikidata.org/wiki/Q577764","display_name":"Communications system","level":2,"score":0.42160001397132874},{"id":"https://openalex.org/C78548338","wikidata":"https://www.wikidata.org/wiki/Q2493","display_name":"Data compression","level":2,"score":0.39259999990463257},{"id":"https://openalex.org/C51632099","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Training set","level":2,"score":0.3856000006198883},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.3716999888420105},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.34689998626708984},{"id":"https://openalex.org/C2777904410","wikidata":"https://www.wikidata.org/wiki/Q7397","display_name":"Software","level":2,"score":0.3465000092983246},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.3434999883174896},{"id":"https://openalex.org/C45374587","wikidata":"https://www.wikidata.org/wiki/Q12525525","display_name":"Computation","level":2,"score":0.313400000333786},{"id":"https://openalex.org/C180016635","wikidata":"https://www.wikidata.org/wiki/Q2712821","display_name":"Compression (physics)","level":2,"score":0.30219998955726624},{"id":"https://openalex.org/C189950617","wikidata":"https://www.wikidata.org/wiki/Q937228","display_name":"Property (philosophy)","level":2,"score":0.3009999990463257},{"id":"https://openalex.org/C199833920","wikidata":"https://www.wikidata.org/wiki/Q612536","display_name":"Vector quantization","level":2,"score":0.29980000853538513},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.2840999960899353},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.27399998903274536},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.27300000190734863},{"id":"https://openalex.org/C25797200","wikidata":"https://www.wikidata.org/wiki/Q828137","display_name":"Compression ratio","level":3,"score":0.2599000036716461},{"id":"https://openalex.org/C2164484","wikidata":"https://www.wikidata.org/wiki/Q5170150","display_name":"Core (optical fiber)","level":2,"score":0.25609999895095825},{"id":"https://openalex.org/C79403827","wikidata":"https://www.wikidata.org/wiki/Q3988","display_name":"Real-time computing","level":1,"score":0.25279998779296875}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/dac63849.2025.11132853","is_oa":false,"landing_page_url":"https://doi.org/10.1109/dac63849.2025.11132853","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 62nd ACM/IEEE Design Automation Conference (DAC)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[{"id":"https://openalex.org/F4320321001","display_name":"National Natural Science Foundation of China","ror":"https://ror.org/01h0zpd94"},{"id":"https://openalex.org/F4320334111","display_name":"Innovation Fund","ror":null}],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":12,"referenced_works":["https://openalex.org/W2618530766","https://openalex.org/W2979515806","https://openalex.org/W3108835732","https://openalex.org/W4220967350","https://openalex.org/W4283809296","https://openalex.org/W4283815817","https://openalex.org/W4385245566","https://openalex.org/W4386078230","https://openalex.org/W4386290345","https://openalex.org/W4386765310","https://openalex.org/W4390097768","https://openalex.org/W4401726194"],"related_works":[],"abstract_inverted_index":{"Mixture-of-Experts":[0],"(MoE)":[1],"model":[2],"parallelism":[3],"is":[4],"prevalent":[5],"in":[6],"training":[7,29,48,79,106,113,132,144],"Large":[8],"Language":[9],"Models":[10],"(e.g.,":[11],"ChatGPT).":[12],"However,":[13],"the":[14,20,68,77,92,104,136,142],"intensive":[15],"all-to-all":[16,100],"collective":[17],"communication":[18,40,93,101,125],"of":[19],"MoE":[21,28,47,78,105,112,143],"layer\u2019s":[22],"intermediate":[23,70],"computing":[24,71],"results":[25,72],"substantially":[26],"degrades":[27],"efficiency.":[30],"In":[31],"this":[32],"paper,":[33],"we":[34],"propose":[35],"BirdMoE,":[36],"a":[37,56,84],"novel":[38],"load-aware":[39],"compression":[41,126,138],"technique":[42],"with":[43,49,61,135],"Bi-random":[44],"quantization":[45],"for":[46,103],"two":[50],"core":[51],"modules.":[52],"Specifically,":[53],"BirdMoE":[54,82,117],"employs":[55],"lightweight":[57],"Random":[58],"Quantization":[59],"(RQ)":[60],"expectation":[62],"invariance":[63],"property":[64],"to":[65,89],"efficiently":[66],"map":[67],"floating-point":[69],"into":[73],"integers":[74],"while":[75,140],"maintaining":[76,141],"quality.":[80,145],"Additionally,":[81],"utilizes":[83],"Mixed":[85],"Precision":[86],"(MP)":[87],"strategy":[88],"dynamically":[90],"balance":[91],"loads":[94],"among":[95],"expert":[96],"nodes,":[97],"significantly":[98],"improving":[99],"efficiency":[102],"system.":[107],"Experiments":[108],"on":[109],"four":[110],"typical":[111],"tasks":[114],"demonstrate":[115],"that":[116],"achieves":[118],"higher":[119],"$4.06":[120],"\\times-":[121],"10.44":[122],"\\times$":[123,131],"total":[124],"ratios":[127],"and":[128],"$1.18":[129],"\\times-5.27":[130],"speedup":[133],"compared":[134],"state-of-the-art":[137],"techniques":[139]},"counts_by_year":[],"updated_date":"2026-04-09T08:11:56.329763","created_date":"2025-10-10T00:00:00"}