{"id":"https://openalex.org/W4319777991","doi":"https://doi.org/10.1109/tcad.2023.3243879","title":"V-LSTM: An Efficient LSTM Accelerator Using Fixed Nonzero-Ratio Viterbi-Based Pruning","display_name":"V-LSTM: An Efficient LSTM Accelerator Using Fixed Nonzero-Ratio Viterbi-Based Pruning","publication_year":2023,"publication_date":"2023-02-09","ids":{"openalex":"https://openalex.org/W4319777991","doi":"https://doi.org/10.1109/tcad.2023.3243879"},"language":"en","primary_location":{"id":"doi:10.1109/tcad.2023.3243879","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tcad.2023.3243879","pdf_url":null,"source":{"id":"https://openalex.org/S100835903","display_name":"IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems","issn_l":"0278-0070","issn":["0278-0070","1937-4151"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319808","host_organization_name":"Institute of Electrical and Electronics Engineers","host_organization_lineage":["https://openalex.org/P4310319808"],"host_organization_lineage_names":["Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems","raw_type":"journal-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/A5101792302","display_name":"Taesu Kim","orcid":"https://orcid.org/0000-0001-7987-527X"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Taesu Kim","raw_affiliation_strings":["SqueezeBits Research, SqueezeBits Inc., Seoul, South Korea"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"SqueezeBits Research, SqueezeBits Inc., Seoul, South Korea","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5023936391","display_name":"Daehyun Ahn","orcid":"https://orcid.org/0000-0001-8050-9963"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Daehyun Ahn","raw_affiliation_strings":["SqueezeBits Research, SqueezeBits Inc., Seoul, South Korea"],"raw_orcid":"https://orcid.org/0000-0001-8050-9963","affiliations":[{"raw_affiliation_string":"SqueezeBits Research, SqueezeBits Inc., Seoul, South Korea","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101652725","display_name":"Dongsoo Lee","orcid":"https://orcid.org/0000-0002-4155-6940"},"institutions":[{"id":"https://openalex.org/I60922564","display_name":"Naver (South Korea)","ror":"https://ror.org/04nzrnx83","country_code":"KR","type":"company","lineage":["https://openalex.org/I60922564"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Dongsoo Lee","raw_affiliation_strings":["Efficient AI, NAVER Cloud, Seoul, South Korea"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Efficient AI, NAVER Cloud, Seoul, South Korea","institution_ids":["https://openalex.org/I60922564"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5003219699","display_name":"Jae\u2010Joon Kim","orcid":"https://orcid.org/0000-0001-5175-8258"},"institutions":[{"id":"https://openalex.org/I139264467","display_name":"Seoul National University","ror":"https://ror.org/04h9pn542","country_code":"KR","type":"education","lineage":["https://openalex.org/I139264467"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Jae-Joon Kim","raw_affiliation_strings":["Department of Electrical and Computer Engineering, Seoul National University, Seoul, South Korea"],"raw_orcid":"https://orcid.org/0000-0001-5175-8258","affiliations":[{"raw_affiliation_string":"Department of Electrical and Computer Engineering, Seoul National University, Seoul, South Korea","institution_ids":["https://openalex.org/I139264467"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":4,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":1.6316,"has_fulltext":false,"cited_by_count":10,"citation_normalized_percentile":{"value":0.86352302,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":94,"max":98},"biblio":{"volume":"42","issue":"10","first_page":"3327","last_page":"3337"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10201","display_name":"Speech Recognition and Synthesis","score":0.9990000128746033,"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/T10201","display_name":"Speech Recognition and Synthesis","score":0.9990000128746033,"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/T12326","display_name":"Network Packet Processing and Optimization","score":0.9976999759674072,"subfield":{"id":"https://openalex.org/subfields/1708","display_name":"Hardware and Architecture"},"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/T11309","display_name":"Music and Audio Processing","score":0.9965000152587891,"subfield":{"id":"https://openalex.org/subfields/1711","display_name":"Signal Processing"},"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":0.8086374998092651},{"id":"https://openalex.org/keywords/field-programmable-gate-array","display_name":"Field-programmable gate array","score":0.49285054206848145},{"id":"https://openalex.org/keywords/pruning","display_name":"Pruning","score":0.48320257663726807},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.45014646649360657},{"id":"https://openalex.org/keywords/computation","display_name":"Computation","score":0.4471222162246704},{"id":"https://openalex.org/keywords/viterbi-algorithm","display_name":"Viterbi algorithm","score":0.42324405908584595},{"id":"https://openalex.org/keywords/convolutional-neural-network","display_name":"Convolutional neural network","score":0.4203363060951233},{"id":"https://openalex.org/keywords/hardware-acceleration","display_name":"Hardware acceleration","score":0.4139578342437744},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.3786170482635498},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.357172429561615},{"id":"https://openalex.org/keywords/computer-hardware","display_name":"Computer hardware","score":0.2921793460845947},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.27942776679992676},{"id":"https://openalex.org/keywords/hidden-markov-model","display_name":"Hidden Markov model","score":0.21008673310279846}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8086374998092651},{"id":"https://openalex.org/C42935608","wikidata":"https://www.wikidata.org/wiki/Q190411","display_name":"Field-programmable gate array","level":2,"score":0.49285054206848145},{"id":"https://openalex.org/C108010975","wikidata":"https://www.wikidata.org/wiki/Q500094","display_name":"Pruning","level":2,"score":0.48320257663726807},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.45014646649360657},{"id":"https://openalex.org/C45374587","wikidata":"https://www.wikidata.org/wiki/Q12525525","display_name":"Computation","level":2,"score":0.4471222162246704},{"id":"https://openalex.org/C60582962","wikidata":"https://www.wikidata.org/wiki/Q83886","display_name":"Viterbi algorithm","level":3,"score":0.42324405908584595},{"id":"https://openalex.org/C81363708","wikidata":"https://www.wikidata.org/wiki/Q17084460","display_name":"Convolutional neural network","level":2,"score":0.4203363060951233},{"id":"https://openalex.org/C13164978","wikidata":"https://www.wikidata.org/wiki/Q600158","display_name":"Hardware acceleration","level":3,"score":0.4139578342437744},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.3786170482635498},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.357172429561615},{"id":"https://openalex.org/C9390403","wikidata":"https://www.wikidata.org/wiki/Q3966","display_name":"Computer hardware","level":1,"score":0.2921793460845947},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.27942776679992676},{"id":"https://openalex.org/C23224414","wikidata":"https://www.wikidata.org/wiki/Q176769","display_name":"Hidden Markov model","level":2,"score":0.21008673310279846},{"id":"https://openalex.org/C86803240","wikidata":"https://www.wikidata.org/wiki/Q420","display_name":"Biology","level":0,"score":0.0},{"id":"https://openalex.org/C6557445","wikidata":"https://www.wikidata.org/wiki/Q173113","display_name":"Agronomy","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/tcad.2023.3243879","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tcad.2023.3243879","pdf_url":null,"source":{"id":"https://openalex.org/S100835903","display_name":"IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems","issn_l":"0278-0070","issn":["0278-0070","1937-4151"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319808","host_organization_name":"Institute of Electrical and Electronics Engineers","host_organization_lineage":["https://openalex.org/P4310319808"],"host_organization_lineage_names":["Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems","raw_type":"journal-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G688674789","display_name":null,"funder_award_id":"2021-0-01343 (10%)","funder_id":"https://openalex.org/F4320321292","funder_display_name":"Seoul National University"},{"id":"https://openalex.org/G8346320104","display_name":null,"funder_award_id":"2022-0-00266","funder_id":"https://openalex.org/F4320322030","funder_display_name":"Ministry of Science, ICT and Future Planning"}],"funders":[{"id":"https://openalex.org/F4320321292","display_name":"Seoul National University","ror":"https://ror.org/04h9pn542"},{"id":"https://openalex.org/F4320322030","display_name":"Ministry of Science, ICT and Future Planning","ror":"https://ror.org/032e49973"}],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":16,"referenced_works":["https://openalex.org/W1635512741","https://openalex.org/W1999085092","https://openalex.org/W2289252105","https://openalex.org/W2585720638","https://openalex.org/W2904773682","https://openalex.org/W2915106038","https://openalex.org/W2940345045","https://openalex.org/W2962820060","https://openalex.org/W4244330903","https://openalex.org/W6687566353","https://openalex.org/W6719700797","https://openalex.org/W6725543821","https://openalex.org/W6726275242","https://openalex.org/W6727690538","https://openalex.org/W6745265922","https://openalex.org/W6748693672"],"related_works":["https://openalex.org/W2062804535","https://openalex.org/W2148710370","https://openalex.org/W2111241003","https://openalex.org/W2281825358","https://openalex.org/W4200391368","https://openalex.org/W2355315220","https://openalex.org/W2210979487","https://openalex.org/W4225949190","https://openalex.org/W2518118925","https://openalex.org/W3159273459"],"abstract_inverted_index":{"Long":[0],"short-term":[1],"memory":[2,54,99],"(LSTM)":[3],"has":[4,71],"been":[5,72],"widely":[6],"adopted":[7],"in":[8,45,120],"tasks":[9],"with":[10,51,106],"sequence":[11],"data,":[12],"such":[13],"as":[14],"speech":[15],"recognition":[16],"and":[17,39,56,65,80,131],"language":[18],"modeling.":[19],"LSTM":[20,47,68,102,147],"brought":[21],"significant":[22],"accuracy":[23,108],"improvement":[24],"by":[25,104],"introducing":[26],"additional":[27,112],"parameters":[28,38],"to":[29,43,61,167],"recurrent":[30],"neural":[31],"network":[32],"(RNN).":[33],"However,":[34],"increasing":[35],"number":[36],"of":[37,67,101,117,158,175],"computations":[40],"also":[41],"led":[42],"inefficiency":[44],"computing":[46],"on":[48,114,150,183],"edge":[49],"devices":[50],"limited":[52],"on-chip":[53],"size":[55],"DRAM":[57],"bandwidth.":[58],"In":[59,84],"order":[60],"reduce":[62,97],"the":[63,90,98,115,123,128,134,142,151,159,164],"latency":[64,172],"energy":[66],"computations,":[69],"there":[70],"a":[73],"pressing":[74],"need":[75],"for":[76,193],"model":[77],"compression":[78,156],"schemes":[79],"suitable":[81],"hardware":[82],"accelerators.":[83,177],"this":[85],"article,":[86],"we":[87,140],"first":[88],"propose":[89,141],"Fixed":[91],"Nonzero-ratio":[92],"Viterbi-based":[93,121],"Pruning,":[94,122],"which":[95],"can":[96],"footprint":[100],"models":[103],"96%":[105],"negligible":[107],"loss.":[109],"By":[110],"applying":[111],"constraints":[113],"distribution":[116],"surviving":[118],"weights":[119],"proposed":[124,152,160,165,179],"pruning":[125,153,161],"scheme":[126,162],"mitigates":[127],"load-imbalance":[129],"problem":[130],"thereby":[132],"increases":[133],"processing":[135],"engine":[136],"utilization":[137],"rate.":[138],"Then,":[139],"V-LSTM,":[143],"an":[144],"efficient":[145],"sparse":[146],"accelerator":[148,166,180],"based":[149],"scheme.":[154],"High":[155],"ratio":[157],"allows":[163],"achieve":[168],"24.9%":[169],"lower":[170],"per-sample":[171],"than":[173],"that":[174],"state-of-the-art":[176],"The":[178],"is":[181],"implemented":[182],"Xilinx":[184],"VC-709":[185],"FPGA":[186],"evaluation":[187],"board":[188],"running":[189],"at":[190],"200":[191],"MHz":[192],"evaluation.":[194]},"counts_by_year":[{"year":2026,"cited_by_count":1},{"year":2025,"cited_by_count":4},{"year":2024,"cited_by_count":3},{"year":2023,"cited_by_count":2}],"updated_date":"2026-06-11T09:08:48.828518","created_date":"2025-10-10T00:00:00"}