{"id":"https://openalex.org/W4283314135","doi":"https://doi.org/10.1109/isqed54688.2022.9806197","title":"An Automatic and Efficient BERT Pruning for Edge AI Systems","display_name":"An Automatic and Efficient BERT Pruning for Edge AI Systems","publication_year":2022,"publication_date":"2022-04-06","ids":{"openalex":"https://openalex.org/W4283314135","doi":"https://doi.org/10.1109/isqed54688.2022.9806197"},"language":"en","primary_location":{"id":"doi:10.1109/isqed54688.2022.9806197","is_oa":false,"landing_page_url":"https://doi.org/10.1109/isqed54688.2022.9806197","pdf_url":null,"source":{"id":"https://openalex.org/S4363607671","display_name":"2022 23rd International Symposium on Quality Electronic Design (ISQED)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2022 23rd International Symposium on Quality Electronic Design (ISQED)","raw_type":"proceedings-article"},"type":"conference-paper","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/A5100529062","display_name":"Shaoyi Huang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Shaoyi Huang","raw_affiliation_strings":["University of Connecticut"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Connecticut","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100432426","display_name":"Ning Liu","orcid":"https://orcid.org/0000-0003-4943-6625"},"institutions":[{"id":"https://openalex.org/I87182695","display_name":"Universidad del Noreste","ror":"https://ror.org/02ahky613","country_code":"MX","type":"education","lineage":["https://openalex.org/I87182695"]}],"countries":["MX"],"is_corresponding":false,"raw_author_name":"Ning Liu","raw_affiliation_strings":["Northeastern University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Northeastern University","institution_ids":["https://openalex.org/I87182695"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5109247936","display_name":"Yueying Liang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yueying Liang","raw_affiliation_strings":["University of Connecticut"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Connecticut","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5045835605","display_name":"Hongwu Peng","orcid":"https://orcid.org/0000-0003-2025-2195"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Hongwu Peng","raw_affiliation_strings":["University of Connecticut"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Connecticut","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101911275","display_name":"Hongjia Li","orcid":"https://orcid.org/0000-0003-1683-343X"},"institutions":[{"id":"https://openalex.org/I87182695","display_name":"Universidad del Noreste","ror":"https://ror.org/02ahky613","country_code":"MX","type":"education","lineage":["https://openalex.org/I87182695"]}],"countries":["MX"],"is_corresponding":false,"raw_author_name":"Hongjia Li","raw_affiliation_strings":["Northeastern University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Northeastern University","institution_ids":["https://openalex.org/I87182695"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5068433690","display_name":"Dongkuan Xu","orcid":"https://orcid.org/0000-0002-1456-9658"},"institutions":[{"id":"https://openalex.org/I130769515","display_name":"Pennsylvania State University","ror":"https://ror.org/04p491231","country_code":"US","type":"education","lineage":["https://openalex.org/I130769515"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Dongkuan Xu","raw_affiliation_strings":["The Pennsylvania State University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"The Pennsylvania State University","institution_ids":["https://openalex.org/I130769515"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5059911353","display_name":"Mimi Xie","orcid":"https://orcid.org/0000-0003-1973-2909"},"institutions":[{"id":"https://openalex.org/I45438204","display_name":"The University of Texas at San Antonio","ror":"https://ror.org/01kd65564","country_code":"US","type":"education","lineage":["https://openalex.org/I45438204"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Mimi Xie","raw_affiliation_strings":["University of Texas at San Antonio"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Texas at San Antonio","institution_ids":["https://openalex.org/I45438204"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5030060072","display_name":"Caiwen Ding","orcid":"https://orcid.org/0000-0003-0891-1231"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Caiwen Ding","raw_affiliation_strings":["University of Connecticut"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Connecticut","institution_ids":[]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":3,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":12,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"6"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.9998999834060669,"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":0.9998999834060669,"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/T10181","display_name":"Natural Language Processing Techniques","score":0.9991000294685364,"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/T11714","display_name":"Multimodal Machine Learning Applications","score":0.996999979019165,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.7795522212982178},{"id":"https://openalex.org/keywords/pruning","display_name":"Pruning","score":0.661385715007782},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.5688446164131165},{"id":"https://openalex.org/keywords/language-model","display_name":"Language model","score":0.5179325938224792},{"id":"https://openalex.org/keywords/deep-learning","display_name":"Deep learning","score":0.48883336782455444},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.4327065944671631},{"id":"https://openalex.org/keywords/encoder","display_name":"Encoder","score":0.4210665822029114}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7795522212982178},{"id":"https://openalex.org/C108010975","wikidata":"https://www.wikidata.org/wiki/Q500094","display_name":"Pruning","level":2,"score":0.661385715007782},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5688446164131165},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.5179325938224792},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.48883336782455444},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.4327065944671631},{"id":"https://openalex.org/C118505674","wikidata":"https://www.wikidata.org/wiki/Q42586063","display_name":"Encoder","level":2,"score":0.4210665822029114},{"id":"https://openalex.org/C6557445","wikidata":"https://www.wikidata.org/wiki/Q173113","display_name":"Agronomy","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/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/isqed54688.2022.9806197","is_oa":false,"landing_page_url":"https://doi.org/10.1109/isqed54688.2022.9806197","pdf_url":null,"source":{"id":"https://openalex.org/S4363607671","display_name":"2022 23rd International Symposium on Quality Electronic Design (ISQED)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2022 23rd International Symposium on Quality Electronic Design (ISQED)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":57,"referenced_works":["https://openalex.org/W131533222","https://openalex.org/W1902237438","https://openalex.org/W2739351760","https://openalex.org/W2805003733","https://openalex.org/W2805249667","https://openalex.org/W2886851211","https://openalex.org/W2892009249","https://openalex.org/W2896457183","https://openalex.org/W2923014074","https://openalex.org/W2962369866","https://openalex.org/W2962785754","https://openalex.org/W2962820060","https://openalex.org/W2970597249","https://openalex.org/W2976833415","https://openalex.org/W2978017171","https://openalex.org/W2995197005","https://openalex.org/W2997768846","https://openalex.org/W3008374555","https://openalex.org/W3015468748","https://openalex.org/W3034445277","https://openalex.org/W3038012435","https://openalex.org/W3094502228","https://openalex.org/W3100980998","https://openalex.org/W3103754749","https://openalex.org/W3104033643","https://openalex.org/W3104136798","https://openalex.org/W3104263050","https://openalex.org/W3105966348","https://openalex.org/W3162542754","https://openalex.org/W3168125510","https://openalex.org/W3174064855","https://openalex.org/W3176468986","https://openalex.org/W3194984065","https://openalex.org/W3205816523","https://openalex.org/W3206837665","https://openalex.org/W3212348580","https://openalex.org/W4287828539","https://openalex.org/W4288086191","https://openalex.org/W4295838474","https://openalex.org/W4385245566","https://openalex.org/W6605323724","https://openalex.org/W6739901393","https://openalex.org/W6751979845","https://openalex.org/W6755207826","https://openalex.org/W6763701032","https://openalex.org/W6765939562","https://openalex.org/W6768429542","https://openalex.org/W6768851824","https://openalex.org/W6771400683","https://openalex.org/W6771626834","https://openalex.org/W6773815586","https://openalex.org/W6774302960","https://openalex.org/W6776048684","https://openalex.org/W6779879114","https://openalex.org/W6780482815","https://openalex.org/W6784333009","https://openalex.org/W6803050810"],"related_works":["https://openalex.org/W2373300491","https://openalex.org/W4375867731","https://openalex.org/W4390516098","https://openalex.org/W2378744544","https://openalex.org/W2594301978","https://openalex.org/W2379704676","https://openalex.org/W2181948922","https://openalex.org/W1998810860","https://openalex.org/W4206442282","https://openalex.org/W4380075502"],"abstract_inverted_index":{"With":[0],"the":[1,74,111,145,148,187,218,222],"yearning":[2],"for":[3,22],"deep":[4],"learning":[5],"democratization,":[6],"there":[7],"are":[8],"increasing":[9],"demands":[10],"to":[11,34,38,64,174,209],"implement":[12],"Transformer-based":[13],"natural":[14],"language":[15],"processing":[16],"(NLP)":[17],"models":[18],"on":[19,93,100,117,198,233],"resource-constrained":[20],"devices":[21],"low-latency":[23],"and":[24,46,56,87,121,126],"high":[25,71],"accuracy.":[26,47,135],"Existing":[27],"BERT":[28,58,193,228],"pruning":[29,59,76,115,130,151,163,177],"methods":[30,116],"require":[31],"domain":[32],"experts":[33,85],"heuristically":[35],"handcraft":[36],"hyperparameters":[37],"strike":[39],"a":[40,66,89,140,160,166,191,205],"balance":[41],"among":[42],"model":[43,185,232],"size,":[44],"latency,":[45],"In":[48],"this":[49],"work,":[50],"we":[51,123,138,157],"propose":[52],"AE-BERT,":[53],"an":[54],"automatic":[55],"efficient":[57,62],"framework":[60],"with":[61,165],"evaluation":[63],"select":[65],"\"good\"":[67],"sub-network":[68],"candidate":[69],"(with":[70],"accuracy)":[72],"given":[73],"overall":[75,129,150],"ratio":[77,131,152,164],"constraints.":[78],"Our":[79,97],"proposed":[80,223],"method":[81,224],"requires":[82],"no":[83],"human":[84],"experience":[86],"achieves":[88],"better":[90],"accuracy":[91],"performance":[92],"many":[94],"NLP":[95],"tasks.":[96],"experimental":[98],"results":[99,180],"General":[101],"Language":[102],"Understanding":[103],"Evaluation":[104],"(GLUE)":[105],"benchmark":[106],"show":[107,182],"that":[108,183],"AE-BERT":[109],"outperforms":[110],"state-of-the-art":[112],"(SOTA)":[113],"hand-crafted":[114,176],"BERT.":[118],"On":[119,136,155],"QNLI":[120],"RTE,":[122],"obtain":[124,139],"75%":[125],"42.8%":[127],"more":[128],"while":[132],"achieving":[133],"higher":[134,142,162],"MRPC,":[137],"4.6":[141],"score":[143],"than":[144],"SOTA":[146,175],"at":[147],"same":[149],"of":[153,190,220,227],"0.5.":[154],"STS-B,":[156],"can":[158],"achieve":[159],"40%":[161],"very":[167],"small":[168],"loss":[169],"in":[170],"Spearman":[171],"correlation":[172],"compared":[173,208],"methods.":[178],"Experimental":[179],"also":[181],"after":[184],"compression,":[186],"inference":[188],"time":[189],"single":[192],"<inf":[194,229],"xmlns:mml=\"http://www.w3.org/1998/Math/MathML\"":[195,230],"xmlns:xlink=\"http://www.w3.org/1999/xlink\">BASE</inf>":[196,231],"encoder":[197],"Xilinx":[199],"Alveo":[200],"U200":[201],"FPGA":[202],"board":[203],"has":[204],"1.83\u00d7":[206],"speedup":[207],"Intel(R)":[210],"Xeon(R)":[211],"Gold":[212],"5218":[213],"(2.30GHz)":[214],"CPU,":[215],"which":[216],"shows":[217],"reasonableness":[219],"deploying":[221],"generated":[225],"sub-networks":[226],"computation":[234],"restricted":[235],"devices.":[236]},"counts_by_year":[{"year":2025,"cited_by_count":6},{"year":2024,"cited_by_count":1},{"year":2023,"cited_by_count":1},{"year":2022,"cited_by_count":4}],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-10-10T00:00:00"}