{"id":"https://openalex.org/W3036463250","doi":"https://doi.org/10.18653/v1/2020.sustainlp-1.17","title":"SqueezeBERT: What can computer vision teach NLP about efficient neural networks?","display_name":"SqueezeBERT: What can computer vision teach NLP about efficient neural networks?","publication_year":2020,"publication_date":"2020-01-01","ids":{"openalex":"https://openalex.org/W3036463250","doi":"https://doi.org/10.18653/v1/2020.sustainlp-1.17","mag":"3036463250"},"language":"en","primary_location":{"id":"doi:10.18653/v1/2020.sustainlp-1.17","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2020.sustainlp-1.17","pdf_url":"https://www.aclweb.org/anthology/2020.sustainlp-1.17.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":"Proceedings of SustaiNLP: Workshop on Simple and Efficient Natural Language Processing","raw_type":"proceedings-article"},"type":"preprint","indexed_in":["arxiv","crossref","datacite"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://www.aclweb.org/anthology/2020.sustainlp-1.17.pdf","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5031485831","display_name":"Forrest Iandola","orcid":null},"institutions":[{"id":"https://openalex.org/I95457486","display_name":"University of California, Berkeley","ror":"https://ror.org/01an7q238","country_code":"US","type":"education","lineage":["https://openalex.org/I95457486"]}],"countries":["US"],"is_corresponding":true,"raw_author_name":"Forrest Iandola","raw_affiliation_strings":["University of California\u2013Berkeley"],"affiliations":[{"raw_affiliation_string":"University of California\u2013Berkeley","institution_ids":["https://openalex.org/I95457486"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5112853966","display_name":"Albert Shaw","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Albert Shaw","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5018488422","display_name":"Ravi Krishna","orcid":null},"institutions":[{"id":"https://openalex.org/I95457486","display_name":"University of California, Berkeley","ror":"https://ror.org/01an7q238","country_code":"US","type":"education","lineage":["https://openalex.org/I95457486"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Ravi Krishna","raw_affiliation_strings":["University of California\u2013Berkeley"],"affiliations":[{"raw_affiliation_string":"University of California\u2013Berkeley","institution_ids":["https://openalex.org/I95457486"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5047285420","display_name":"Kurt Keutzer","orcid":"https://orcid.org/0000-0003-3868-8501"},"institutions":[{"id":"https://openalex.org/I95457486","display_name":"University of California, Berkeley","ror":"https://ror.org/01an7q238","country_code":"US","type":"education","lineage":["https://openalex.org/I95457486"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Kurt Keutzer","raw_affiliation_strings":["University of California\u2013Berkeley"],"affiliations":[{"raw_affiliation_string":"University of California\u2013Berkeley","institution_ids":["https://openalex.org/I95457486"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":4,"corresponding_author_ids":["https://openalex.org/A5031485831"],"corresponding_institution_ids":["https://openalex.org/I95457486"],"apc_list":null,"apc_paid":null,"fwci":2.09934636,"has_fulltext":true,"cited_by_count":20,"citation_normalized_percentile":{"value":0.88474195,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":94,"max":99},"biblio":{"volume":null,"issue":null,"first_page":"124","last_page":"135"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.9995999932289124,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"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/T10036","display_name":"Advanced Neural Network Applications","score":0.9995999932289124,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"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/T11714","display_name":"Multimodal Machine Learning Applications","score":0.9994999766349792,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"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/T11512","display_name":"Anomaly Detection Techniques and Applications","score":0.9987000226974487,"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":0.8529442548751831},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.6655447483062744},{"id":"https://openalex.org/keywords/set","display_name":"Set (abstract data type)","score":0.5901708602905273},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.5870410203933716},{"id":"https://openalex.org/keywords/snippet","display_name":"Snippet","score":0.521485447883606},{"id":"https://openalex.org/keywords/test-set","display_name":"Test set","score":0.502406120300293},{"id":"https://openalex.org/keywords/architecture","display_name":"Architecture","score":0.4954870939254761},{"id":"https://openalex.org/keywords/deep-learning","display_name":"Deep learning","score":0.47372475266456604},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.45875975489616394},{"id":"https://openalex.org/keywords/code","display_name":"Code (set theory)","score":0.4401858150959015},{"id":"https://openalex.org/keywords/base","display_name":"Base (topology)","score":0.41231200098991394},{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.3419809341430664},{"id":"https://openalex.org/keywords/information-retrieval","display_name":"Information retrieval","score":0.21284925937652588},{"id":"https://openalex.org/keywords/programming-language","display_name":"Programming language","score":0.12983307242393494}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8529442548751831},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6655447483062744},{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"score":0.5901708602905273},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.5870410203933716},{"id":"https://openalex.org/C2777822670","wikidata":"https://www.wikidata.org/wiki/Q1120538","display_name":"Snippet","level":2,"score":0.521485447883606},{"id":"https://openalex.org/C169903167","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Test set","level":2,"score":0.502406120300293},{"id":"https://openalex.org/C123657996","wikidata":"https://www.wikidata.org/wiki/Q12271","display_name":"Architecture","level":2,"score":0.4954870939254761},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.47372475266456604},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.45875975489616394},{"id":"https://openalex.org/C2776760102","wikidata":"https://www.wikidata.org/wiki/Q5139990","display_name":"Code (set theory)","level":3,"score":0.4401858150959015},{"id":"https://openalex.org/C42058472","wikidata":"https://www.wikidata.org/wiki/Q810214","display_name":"Base (topology)","level":2,"score":0.41231200098991394},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.3419809341430664},{"id":"https://openalex.org/C23123220","wikidata":"https://www.wikidata.org/wiki/Q816826","display_name":"Information retrieval","level":1,"score":0.21284925937652588},{"id":"https://openalex.org/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.12983307242393494},{"id":"https://openalex.org/C153349607","wikidata":"https://www.wikidata.org/wiki/Q36649","display_name":"Visual arts","level":1,"score":0.0},{"id":"https://openalex.org/C134306372","wikidata":"https://www.wikidata.org/wiki/Q7754","display_name":"Mathematical analysis","level":1,"score":0.0},{"id":"https://openalex.org/C142362112","wikidata":"https://www.wikidata.org/wiki/Q735","display_name":"Art","level":0,"score":0.0},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.0}],"mesh":[],"locations_count":4,"locations":[{"id":"doi:10.18653/v1/2020.sustainlp-1.17","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2020.sustainlp-1.17","pdf_url":"https://www.aclweb.org/anthology/2020.sustainlp-1.17.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":"Proceedings of SustaiNLP: Workshop on Simple and Efficient Natural Language Processing","raw_type":"proceedings-article"},{"id":"pmh:oai:arXiv.org:2006.11316","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2006.11316","pdf_url":"https://arxiv.org/pdf/2006.11316","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"},{"id":"mag:3036463250","is_oa":true,"landing_page_url":"https://arxiv.org/pdf/2006.11316.pdf","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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"arXiv (Cornell University)","raw_type":null},{"id":"doi:10.48550/arxiv.2006.11316","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2006.11316","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":"doi:10.18653/v1/2020.sustainlp-1.17","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2020.sustainlp-1.17","pdf_url":"https://www.aclweb.org/anthology/2020.sustainlp-1.17.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":"Proceedings of SustaiNLP: Workshop on Simple and Efficient Natural Language Processing","raw_type":"proceedings-article"},"sustainable_development_goals":[{"score":0.47999998927116394,"id":"https://metadata.un.org/sdg/4","display_name":"Quality Education"}],"awards":[],"funders":[{"id":"https://openalex.org/F4320332195","display_name":"Samsung","ror":"https://ror.org/04w3jy968"}],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W3036463250.pdf","grobid_xml":"https://content.openalex.org/works/W3036463250.grobid-xml"},"referenced_works_count":77,"referenced_works":["https://openalex.org/W131533222","https://openalex.org/W1566289585","https://openalex.org/W1599016936","https://openalex.org/W1686810756","https://openalex.org/W1821462560","https://openalex.org/W1832693441","https://openalex.org/W1861492603","https://openalex.org/W1901129140","https://openalex.org/W1903029394","https://openalex.org/W2101926813","https://openalex.org/W2109553965","https://openalex.org/W2117539524","https://openalex.org/W2121879602","https://openalex.org/W2130158090","https://openalex.org/W2147800946","https://openalex.org/W2163605009","https://openalex.org/W2194775991","https://openalex.org/W2251939518","https://openalex.org/W2279098554","https://openalex.org/W2340897893","https://openalex.org/W2396767181","https://openalex.org/W2525127255","https://openalex.org/W2525778437","https://openalex.org/W2531409750","https://openalex.org/W2549139847","https://openalex.org/W2565639579","https://openalex.org/W2612445135","https://openalex.org/W2799054028","https://openalex.org/W2898700502","https://openalex.org/W2908336025","https://openalex.org/W2908510526","https://openalex.org/W2912521296","https://openalex.org/W2945290257","https://openalex.org/W2950179405","https://openalex.org/W2950888501","https://openalex.org/W2952122856","https://openalex.org/W2955425717","https://openalex.org/W2962911728","https://openalex.org/W2963125010","https://openalex.org/W2963341956","https://openalex.org/W2963403868","https://openalex.org/W2963446712","https://openalex.org/W2963748441","https://openalex.org/W2963816728","https://openalex.org/W2963840672","https://openalex.org/W2963846996","https://openalex.org/W2963854351","https://openalex.org/W2964309882","https://openalex.org/W2965373594","https://openalex.org/W2967733054","https://openalex.org/W2970454332","https://openalex.org/W2970583189","https://openalex.org/W2970971581","https://openalex.org/W2971307358","https://openalex.org/W2973049837","https://openalex.org/W2974875810","https://openalex.org/W2975429091","https://openalex.org/W2978017171","https://openalex.org/W2978670439","https://openalex.org/W2979314664","https://openalex.org/W2980282514","https://openalex.org/W2981701755","https://openalex.org/W2986591322","https://openalex.org/W2995435108","https://openalex.org/W2996035354","https://openalex.org/W2996428491","https://openalex.org/W2998183051","https://openalex.org/W3005714399","https://openalex.org/W3012324658","https://openalex.org/W3023547440","https://openalex.org/W3024171804","https://openalex.org/W3030163527","https://openalex.org/W3032388710","https://openalex.org/W3034457371","https://openalex.org/W3034971973","https://openalex.org/W3101248447","https://openalex.org/W3104033643"],"related_works":["https://openalex.org/W2978017171","https://openalex.org/W3105645800","https://openalex.org/W2963341956","https://openalex.org/W2963310665","https://openalex.org/W3034457371","https://openalex.org/W2996428491","https://openalex.org/W2963403868","https://openalex.org/W2965373594","https://openalex.org/W2924902521","https://openalex.org/W1821462560","https://openalex.org/W2415378597","https://openalex.org/W3005049034","https://openalex.org/W2251202267","https://openalex.org/W2603037274","https://openalex.org/W2975608192","https://openalex.org/W2731619549","https://openalex.org/W2905173788","https://openalex.org/W3108426037","https://openalex.org/W2995760626","https://openalex.org/W2797810576"],"abstract_inverted_index":{"Humans":[0],"read":[1],"and":[2,22,39,61,68,91,158],"write":[3],"hundreds":[4],"of":[5,7,16,133],"billions":[6],"messages":[8],"every":[9],"day.":[10],"Further,":[11],"due":[12],"to":[13,49,55,102,148],"the":[14,177,185],"availability":[15],"large":[17,19],"datasets,":[18],"computing":[20],"systems,":[21],"better":[23],"neural":[24,85,142],"network":[25,86,143,166],"models,":[26],"natural":[27],"language":[28],"processing":[29],"(NLP)":[30],"technology":[31],"has":[32],"made":[33],"significant":[34,47,125],"strides":[35],"in":[36,52,152,163],"understanding,":[37],"proofreading,":[38],"organizing":[40],"these":[41,134],"messages.":[42],"Thus,":[43],"there":[44],"is":[45],"a":[46,104,108,164],"opportunity":[48],"deploy":[50],"NLP":[51,77,84,141],"myriad":[53],"applications":[54],"help":[56],"web":[57],"users,":[58],"social":[59],"networks,":[60,130],"businesses.":[62],"In":[63,112],"particular,":[64],"we":[65,115,159],"consider":[66],"smartphones":[67],"other":[69],"mobile":[70],"devices":[71],"as":[72,89,120],"crucial":[73],"platforms":[74],"for":[75,127],"deploying":[76],"models":[78,87],"at":[79],"scale.":[80],"However,":[81],"today's":[82],"highly-accurate":[83],"such":[88,119],"BERT":[90],"RoBERTa":[92],"are":[93],"extremely":[94],"computationally":[95],"expensive,":[96],"with":[97,155],"BERT-base":[98,175],"taking":[99],"1.7":[100],"seconds":[101],"classify":[103],"text":[105],"snippet":[106],"on":[107,176,184],"Pixel":[109,178],"3":[110,179],"smartphone.":[111],"this":[113,161],"work,":[114],"observe":[116],"that":[117],"methods":[118],"grouped":[121,156],"convolutions":[122],"have":[123,136],"yielded":[124],"speedups":[126],"computer":[128],"vision":[129],"but":[131],"many":[132],"techniques":[135],"not":[137],"been":[138],"adopted":[139],"by":[140],"designers.":[144],"We":[145],"demonstrate":[146],"how":[147],"replace":[149],"several":[150],"operations":[151],"self-attention":[153],"layers":[154],"convolutions,":[157],"use":[160],"technique":[162],"novel":[165],"architecture":[167],"called":[168],"SqueezeBERT,":[169],"which":[170],"runs":[171],"4.3x":[172],"faster":[173],"than":[174],"while":[180],"achieving":[181],"competitive":[182],"accuracy":[183],"GLUE":[186],"test":[187],"set.":[188],"The":[189],"SqueezeBERT":[190],"code":[191],"will":[192],"be":[193],"released.":[194]},"counts_by_year":[{"year":2023,"cited_by_count":2},{"year":2022,"cited_by_count":4},{"year":2021,"cited_by_count":10},{"year":2020,"cited_by_count":4}],"updated_date":"2026-02-09T09:26:11.010843","created_date":"2025-10-10T00:00:00"}