{"id":"https://openalex.org/W4294294851","doi":"https://doi.org/10.48550/arxiv.2208.14580","title":"Efficient Sparsely Activated Transformers","display_name":"Efficient Sparsely Activated Transformers","publication_year":2022,"publication_date":"2022-08-31","ids":{"openalex":"https://openalex.org/W4294294851","doi":"https://doi.org/10.48550/arxiv.2208.14580"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2208.14580","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2208.14580","pdf_url":"https://arxiv.org/pdf/2208.14580","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":"","raw_type":null},"type":"preprint","indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2208.14580","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5066128966","display_name":"Salar Latifi","orcid":null},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Latifi, Salar","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5033185286","display_name":"Saurav Muralidharan","orcid":"https://orcid.org/0000-0003-4024-3958"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Muralidharan, Saurav","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5024606205","display_name":"Michael Garland","orcid":"https://orcid.org/0000-0001-6093-7602"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Garland, Michael","raw_affiliation_strings":[],"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":3,"corresponding_author_ids":["https://openalex.org/A5066128966"],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":2,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":null,"last_page":null},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.9926999807357788,"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"}},"topics":[{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.9926999807357788,"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/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.9916999936103821,"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/T10320","display_name":"Neural Networks and Applications","score":0.9897000193595886,"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/inference","display_name":"Inference","score":0.7597109079360962},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.7588639855384827},{"id":"https://openalex.org/keywords/transformer","display_name":"Transformer","score":0.7323821783065796},{"id":"https://openalex.org/keywords/latency","display_name":"Latency (audio)","score":0.5591144561767578},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.5294843316078186},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.4501681923866272},{"id":"https://openalex.org/keywords/computer-engineering","display_name":"Computer engineering","score":0.3962639272212982},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.33492013812065125},{"id":"https://openalex.org/keywords/engineering","display_name":"Engineering","score":0.15761640667915344},{"id":"https://openalex.org/keywords/voltage","display_name":"Voltage","score":0.11381959915161133},{"id":"https://openalex.org/keywords/telecommunications","display_name":"Telecommunications","score":0.0946677029132843},{"id":"https://openalex.org/keywords/electrical-engineering","display_name":"Electrical engineering","score":0.08794751763343811}],"concepts":[{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.7597109079360962},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7588639855384827},{"id":"https://openalex.org/C66322947","wikidata":"https://www.wikidata.org/wiki/Q11658","display_name":"Transformer","level":3,"score":0.7323821783065796},{"id":"https://openalex.org/C82876162","wikidata":"https://www.wikidata.org/wiki/Q17096504","display_name":"Latency (audio)","level":2,"score":0.5591144561767578},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.5294843316078186},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4501681923866272},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.3962639272212982},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.33492013812065125},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.15761640667915344},{"id":"https://openalex.org/C165801399","wikidata":"https://www.wikidata.org/wiki/Q25428","display_name":"Voltage","level":2,"score":0.11381959915161133},{"id":"https://openalex.org/C76155785","wikidata":"https://www.wikidata.org/wiki/Q418","display_name":"Telecommunications","level":1,"score":0.0946677029132843},{"id":"https://openalex.org/C119599485","wikidata":"https://www.wikidata.org/wiki/Q43035","display_name":"Electrical engineering","level":1,"score":0.08794751763343811}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2208.14580","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2208.14580","pdf_url":"https://arxiv.org/pdf/2208.14580","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":"","raw_type":null},{"id":"doi:10.48550/arxiv.2208.14580","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2208.14580","pdf_url":null,"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":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"article"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2208.14580","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2208.14580","pdf_url":"https://arxiv.org/pdf/2208.14580","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":"","raw_type":null},"sustainable_development_goals":[{"score":0.6000000238418579,"display_name":"Quality Education","id":"https://metadata.un.org/sdg/4"}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W2055243143","https://openalex.org/W4321636575","https://openalex.org/W3008625068","https://openalex.org/W3128807919","https://openalex.org/W3176411177","https://openalex.org/W4206178588","https://openalex.org/W4287635093","https://openalex.org/W3094491777","https://openalex.org/W3214715529","https://openalex.org/W3035501883"],"abstract_inverted_index":{"Transformer-based":[0,74],"neural":[1],"networks":[2,38],"have":[3],"achieved":[4],"state-of-the-art":[5],"task":[6],"performance":[7],"in":[8,39],"a":[9,58,65,77],"number":[10],"of":[11,33,42,53,87,120],"machine":[12],"learning":[13],"domains":[14],"including":[15],"natural":[16],"language":[17,108],"processing":[18],"and":[19,76,81,115],"computer":[20],"vision.":[21],"To":[22],"further":[23],"improve":[24],"their":[25],"accuracy,":[26],"recent":[27],"work":[28],"has":[29],"explored":[30],"the":[31,40,51,88,95,112],"integration":[32],"dynamic":[34],"behavior":[35],"into":[36],"these":[37],"form":[41],"mixture-of-expert":[43],"(MoE)":[44],"layers.":[45],"In":[46],"this":[47],"paper,":[48],"we":[49],"explore":[50],"introduction":[52],"MoE":[54],"layers":[55],"to":[56,93],"optimize":[57],"different":[59],"metric:":[60],"inference":[61,117],"latency.":[62],"We":[63,102],"introduce":[64],"novel":[66],"system":[67],"named":[68],"PLANER":[69,104],"that":[70,91],"takes":[71],"an":[72,83],"existing":[73],"network":[75,90,114],"user-defined":[78],"latency":[79,96,118],"target":[80,97],"produces":[82],"optimized,":[84],"sparsely-activated":[85],"version":[86],"original":[89],"tries":[92],"meet":[94],"while":[98],"maintaining":[99],"baseline":[100],"accuracy.":[101],"evaluate":[103],"on":[105],"two":[106],"real-world":[107],"modeling":[109],"tasks":[110],"using":[111],"Transformer-XL":[113],"achieve":[116],"reductions":[119],"over":[121],"2x":[122],"at":[123],"iso-accuracy.":[124]},"counts_by_year":[{"year":2023,"cited_by_count":2}],"updated_date":"2025-11-06T06:51:31.235846","created_date":"2025-10-10T00:00:00"}