{"id":"https://openalex.org/W4389166750","doi":"https://doi.org/10.1109/iccad57390.2023.10323775","title":"TSTC: Two-Level Sparsity Tensor Core Enabling both Algorithm Flexibility and Hardware Efficiency","display_name":"TSTC: Two-Level Sparsity Tensor Core Enabling both Algorithm Flexibility and Hardware Efficiency","publication_year":2023,"publication_date":"2023-10-28","ids":{"openalex":"https://openalex.org/W4389166750","doi":"https://doi.org/10.1109/iccad57390.2023.10323775"},"language":"en","primary_location":{"id":"doi:10.1109/iccad57390.2023.10323775","is_oa":false,"landing_page_url":"https://doi.org/10.1109/iccad57390.2023.10323775","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2023 IEEE/ACM International Conference on Computer Aided Design (ICCAD)","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/A5100358902","display_name":"Jun Liu","orcid":"https://orcid.org/0000-0002-9663-3317"},"institutions":[{"id":"https://openalex.org/I99065089","display_name":"Tsinghua University","ror":"https://ror.org/03cve4549","country_code":"CN","type":"education","lineage":["https://openalex.org/I99065089"]},{"id":"https://openalex.org/I183067930","display_name":"Shanghai Jiao Tong University","ror":"https://ror.org/0220qvk04","country_code":"CN","type":"education","lineage":["https://openalex.org/I183067930"]}],"countries":["CN"],"is_corresponding":true,"raw_author_name":"Jun Liu","raw_affiliation_strings":["Tsinghua University,BNRist,Dept. of EE","Shanghai Jiao Tong University","Dept. of EE, BNRist, Tsinghua University"],"affiliations":[{"raw_affiliation_string":"Tsinghua University,BNRist,Dept. of EE","institution_ids":["https://openalex.org/I99065089"]},{"raw_affiliation_string":"Shanghai Jiao Tong University","institution_ids":["https://openalex.org/I183067930"]},{"raw_affiliation_string":"Dept. of EE, BNRist, Tsinghua University","institution_ids":["https://openalex.org/I99065089"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5015946486","display_name":"Guohao Dai","orcid":"https://orcid.org/0000-0003-0849-3252"},"institutions":[{"id":"https://openalex.org/I183067930","display_name":"Shanghai Jiao Tong University","ror":"https://ror.org/0220qvk04","country_code":"CN","type":"education","lineage":["https://openalex.org/I183067930"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Guohao Dai","raw_affiliation_strings":["Shanghai Jiao Tong University"],"affiliations":[{"raw_affiliation_string":"Shanghai Jiao Tong University","institution_ids":["https://openalex.org/I183067930"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5003416105","display_name":"Hao Xia","orcid":"https://orcid.org/0000-0001-6817-1779"},"institutions":[{"id":"https://openalex.org/I183067930","display_name":"Shanghai Jiao Tong University","ror":"https://ror.org/0220qvk04","country_code":"CN","type":"education","lineage":["https://openalex.org/I183067930"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Hao Xia","raw_affiliation_strings":["Shanghai Jiao Tong University"],"affiliations":[{"raw_affiliation_string":"Shanghai Jiao Tong University","institution_ids":["https://openalex.org/I183067930"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5102924223","display_name":"Lidong Guo","orcid":"https://orcid.org/0000-0003-2520-4416"},"institutions":[{"id":"https://openalex.org/I99065089","display_name":"Tsinghua University","ror":"https://ror.org/03cve4549","country_code":"CN","type":"education","lineage":["https://openalex.org/I99065089"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Lidong Guo","raw_affiliation_strings":["Tsinghua University,BNRist,Dept. of EE","Dept. of EE, BNRist, Tsinghua University"],"affiliations":[{"raw_affiliation_string":"Tsinghua University,BNRist,Dept. of EE","institution_ids":["https://openalex.org/I99065089"]},{"raw_affiliation_string":"Dept. of EE, BNRist, Tsinghua University","institution_ids":["https://openalex.org/I99065089"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5015425741","display_name":"Xiangsheng Shi","orcid":null},"institutions":[{"id":"https://openalex.org/I99065089","display_name":"Tsinghua University","ror":"https://ror.org/03cve4549","country_code":"CN","type":"education","lineage":["https://openalex.org/I99065089"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xiangsheng Shi","raw_affiliation_strings":["Tsinghua University,BNRist,Dept. of EE","Dept. of EE, BNRist, Tsinghua University"],"affiliations":[{"raw_affiliation_string":"Tsinghua University,BNRist,Dept. of EE","institution_ids":["https://openalex.org/I99065089"]},{"raw_affiliation_string":"Dept. of EE, BNRist, Tsinghua University","institution_ids":["https://openalex.org/I99065089"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101747914","display_name":"Jiaming Xu","orcid":null},"institutions":[{"id":"https://openalex.org/I149594827","display_name":"Xidian University","ror":"https://ror.org/05s92vm98","country_code":"CN","type":"education","lineage":["https://openalex.org/I149594827"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Jiaming Xu","raw_affiliation_strings":["Xidian University"],"affiliations":[{"raw_affiliation_string":"Xidian University","institution_ids":["https://openalex.org/I149594827"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5023755254","display_name":"Huazhong Yang","orcid":"https://orcid.org/0000-0003-2421-353X"},"institutions":[{"id":"https://openalex.org/I99065089","display_name":"Tsinghua University","ror":"https://ror.org/03cve4549","country_code":"CN","type":"education","lineage":["https://openalex.org/I99065089"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Huazhong Yang","raw_affiliation_strings":["Tsinghua University,BNRist,Dept. of EE","Dept. of EE, BNRist, Tsinghua University"],"affiliations":[{"raw_affiliation_string":"Tsinghua University,BNRist,Dept. of EE","institution_ids":["https://openalex.org/I99065089"]},{"raw_affiliation_string":"Dept. of EE, BNRist, Tsinghua University","institution_ids":["https://openalex.org/I99065089"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5100445061","display_name":"Yu Wang","orcid":"https://orcid.org/0000-0001-6108-5157"},"institutions":[{"id":"https://openalex.org/I99065089","display_name":"Tsinghua University","ror":"https://ror.org/03cve4549","country_code":"CN","type":"education","lineage":["https://openalex.org/I99065089"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yu Wang","raw_affiliation_strings":["Tsinghua University,BNRist,Dept. of EE","Dept. of EE, BNRist, Tsinghua University"],"affiliations":[{"raw_affiliation_string":"Tsinghua University,BNRist,Dept. of EE","institution_ids":["https://openalex.org/I99065089"]},{"raw_affiliation_string":"Dept. of EE, BNRist, Tsinghua University","institution_ids":["https://openalex.org/I99065089"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":8,"corresponding_author_ids":["https://openalex.org/A5100358902"],"corresponding_institution_ids":["https://openalex.org/I183067930","https://openalex.org/I99065089"],"apc_list":null,"apc_paid":null,"fwci":2.1235,"has_fulltext":false,"cited_by_count":7,"citation_normalized_percentile":{"value":0.87256098,"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":"1","last_page":"9"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.9945999979972839,"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"}},"topics":[{"id":"https://openalex.org/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.9945999979972839,"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/T12303","display_name":"Tensor decomposition and applications","score":0.9923999905586243,"subfield":{"id":"https://openalex.org/subfields/2605","display_name":"Computational Mathematics"},"field":{"id":"https://openalex.org/fields/26","display_name":"Mathematics"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T11612","display_name":"Stochastic Gradient Optimization Techniques","score":0.9843000173568726,"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":0.8486120700836182},{"id":"https://openalex.org/keywords/flexibility","display_name":"Flexibility (engineering)","score":0.7045447826385498},{"id":"https://openalex.org/keywords/sparse-matrix","display_name":"Sparse matrix","score":0.606620192527771},{"id":"https://openalex.org/keywords/reduction","display_name":"Reduction (mathematics)","score":0.5060452818870544},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.4881316125392914},{"id":"https://openalex.org/keywords/tensor","display_name":"Tensor (intrinsic definition)","score":0.48737627267837524},{"id":"https://openalex.org/keywords/inefficiency","display_name":"Inefficiency","score":0.46656978130340576},{"id":"https://openalex.org/keywords/matrix-multiplication","display_name":"Matrix multiplication","score":0.4425602853298187},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.4417884945869446},{"id":"https://openalex.org/keywords/speedup","display_name":"Speedup","score":0.4129834771156311},{"id":"https://openalex.org/keywords/computer-engineering","display_name":"Computer engineering","score":0.3527145981788635}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8486120700836182},{"id":"https://openalex.org/C2780598303","wikidata":"https://www.wikidata.org/wiki/Q65921492","display_name":"Flexibility (engineering)","level":2,"score":0.7045447826385498},{"id":"https://openalex.org/C56372850","wikidata":"https://www.wikidata.org/wiki/Q1050404","display_name":"Sparse matrix","level":3,"score":0.606620192527771},{"id":"https://openalex.org/C111335779","wikidata":"https://www.wikidata.org/wiki/Q3454686","display_name":"Reduction (mathematics)","level":2,"score":0.5060452818870544},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.4881316125392914},{"id":"https://openalex.org/C155281189","wikidata":"https://www.wikidata.org/wiki/Q3518150","display_name":"Tensor (intrinsic definition)","level":2,"score":0.48737627267837524},{"id":"https://openalex.org/C2778869765","wikidata":"https://www.wikidata.org/wiki/Q6028363","display_name":"Inefficiency","level":2,"score":0.46656978130340576},{"id":"https://openalex.org/C17349429","wikidata":"https://www.wikidata.org/wiki/Q1049914","display_name":"Matrix multiplication","level":3,"score":0.4425602853298187},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.4417884945869446},{"id":"https://openalex.org/C68339613","wikidata":"https://www.wikidata.org/wiki/Q1549489","display_name":"Speedup","level":2,"score":0.4129834771156311},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.3527145981788635},{"id":"https://openalex.org/C202444582","wikidata":"https://www.wikidata.org/wiki/Q837863","display_name":"Pure mathematics","level":1,"score":0.0},{"id":"https://openalex.org/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","level":0,"score":0.0},{"id":"https://openalex.org/C2524010","wikidata":"https://www.wikidata.org/wiki/Q8087","display_name":"Geometry","level":1,"score":0.0},{"id":"https://openalex.org/C162324750","wikidata":"https://www.wikidata.org/wiki/Q8134","display_name":"Economics","level":0,"score":0.0},{"id":"https://openalex.org/C105795698","wikidata":"https://www.wikidata.org/wiki/Q12483","display_name":"Statistics","level":1,"score":0.0},{"id":"https://openalex.org/C175444787","wikidata":"https://www.wikidata.org/wiki/Q39072","display_name":"Microeconomics","level":1,"score":0.0},{"id":"https://openalex.org/C62520636","wikidata":"https://www.wikidata.org/wiki/Q944","display_name":"Quantum mechanics","level":1,"score":0.0},{"id":"https://openalex.org/C84114770","wikidata":"https://www.wikidata.org/wiki/Q46344","display_name":"Quantum","level":2,"score":0.0},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.0},{"id":"https://openalex.org/C163716315","wikidata":"https://www.wikidata.org/wiki/Q901177","display_name":"Gaussian","level":2,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/iccad57390.2023.10323775","is_oa":false,"landing_page_url":"https://doi.org/10.1109/iccad57390.2023.10323775","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2023 IEEE/ACM International Conference on Computer Aided Design (ICCAD)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[{"display_name":"Affordable and clean energy","id":"https://metadata.un.org/sdg/7","score":0.8999999761581421}],"awards":[{"id":"https://openalex.org/G956202704","display_name":null,"funder_award_id":"62104128,U19B2019,61832007","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"}],"funders":[{"id":"https://openalex.org/F4320321001","display_name":"National Natural Science Foundation of China","ror":"https://ror.org/01h0zpd94"}],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":40,"referenced_works":["https://openalex.org/W1530262073","https://openalex.org/W1686810756","https://openalex.org/W2022322548","https://openalex.org/W2027731328","https://openalex.org/W2108598243","https://openalex.org/W2153959628","https://openalex.org/W2168190036","https://openalex.org/W2194775991","https://openalex.org/W2475334473","https://openalex.org/W2507009361","https://openalex.org/W2606722458","https://openalex.org/W2896457183","https://openalex.org/W2953384591","https://openalex.org/W2980186997","https://openalex.org/W3016542674","https://openalex.org/W3016735325","https://openalex.org/W3130554079","https://openalex.org/W3130716829","https://openalex.org/W3132695675","https://openalex.org/W3159727696","https://openalex.org/W3187908937","https://openalex.org/W3198975860","https://openalex.org/W4240168186","https://openalex.org/W4287748555","https://openalex.org/W4288083766","https://openalex.org/W4292779060","https://openalex.org/W4293024985","https://openalex.org/W4294558607","https://openalex.org/W4295312788","https://openalex.org/W4321637181","https://openalex.org/W4360831960","https://openalex.org/W4360831975","https://openalex.org/W6631660994","https://openalex.org/W6637373629","https://openalex.org/W6713134421","https://openalex.org/W6738964360","https://openalex.org/W6755207826","https://openalex.org/W6766978945","https://openalex.org/W6778883912","https://openalex.org/W6780069040"],"related_works":["https://openalex.org/W2293771254","https://openalex.org/W2039875226","https://openalex.org/W4221142455","https://openalex.org/W3121828480","https://openalex.org/W2914631005","https://openalex.org/W2123154672","https://openalex.org/W2032786851","https://openalex.org/W2952630098","https://openalex.org/W3132517792","https://openalex.org/W4285148873"],"abstract_inverted_index":{"The":[0,39],"tensor":[1,41,257],"cores":[2],"in":[3,11,19,29,33,100],"modern":[4],"GPUs":[5],"lead":[6],"to":[7,46,54,80,166,168,211,236,239],"significant":[8],"performance":[9],"improvement":[10],"matrix":[12],"multiplication,":[13],"which":[14,110,124,227],"is":[15],"the":[16,85,107,118,132,195,199,224,256,260],"primary":[17],"operation":[18],"deep":[20,30,233],"learning.":[21],"However,":[22],"existing":[23],"hardware":[24,37,81,91,122,177],"architectures":[25,61],"face":[26],"unstructured":[27,76,108,147],"sparsity":[28,109,127],"learning,":[31],"resulting":[32],"algorithm":[34,55,88,112,136],"inflexibility":[35,89],"and":[36,90,249,259],"inefficiency.":[38,82],"previous":[40],"core":[42,258],"architecture":[43],"requires":[44,125],"matrices":[45,149],"be":[47,115,129,151],"pruned":[48],"into":[49,153],"2:4":[50],"sparse":[51,77,148,190,232],"patterns,":[52],"leading":[53,79],"inflexibility.":[56],"Customized":[57],"accelerators":[58],"introduce":[59],"extra":[60],"(e.g.,":[62],"interconnection":[63],"networks":[64],"for":[65,71,75],"dynamic":[66],"data":[67,73,196],"routing":[68],"or":[69,206],"buffers":[70],"avoiding":[72],"conflicts)":[74],"matrices,":[78],"To":[83],"tackle":[84],"contradiction":[86],"between":[87],"inefficiency,":[92],"we":[93,138,179],"propose":[94,139,180,223],"Two-level":[95],"Sparsity":[96],"Tensor":[97],"Core":[98],"(TSTC)":[99],"this":[101],"paper.":[102],"TSTC":[103,210,242],"points":[104],"out":[105],"that":[106],"enables":[111,146,187,209],"flexibility":[113],"can":[114,128,150,228],"maintained":[116],"at":[117,131],"coarse-grained":[119],"level,":[120],"while":[121],"efficiency":[123],"structured":[126,158,189],"ensured":[130],"fine-grained":[133,154],"level.":[134],"For":[135,176],"flexibility,":[137],"Flexible":[140],"Sparse":[141],"Block":[142],"(FSB)":[143],"pattern.":[144],"FSB":[145,164],"divided":[152],"blocks":[155],"with":[156,173],"different":[157,188,231],"sparsity.":[159],"As":[160],"a":[161,218],"result,":[162],"using":[163],"leads":[165],"up":[167,171],"7.29x":[169],"speed":[170],"compared":[172],"other":[174],"formats.":[175],"efficiency,":[178],"Dynamic":[181],"Extendible":[182],"Reduction":[183],"Network":[184],"(DERN).":[185],"DERN":[186,208],"reductions":[191],"by":[192],"only":[193],"extending":[194],"width":[197],"on":[198],"standard":[200],"reduction":[201],"network":[202],"without":[203],"introducing":[204],"interconnections":[205],"buffers.":[207],"achieve":[212],"7.19x":[213],"more":[214],"energy":[215,253],"savings":[216,254],"under":[217],"similar":[219],"speed.":[220],"We":[221],"also":[222],"whole":[225],"flow,":[226],"automatically":[229],"deploy":[230],"learning":[234],"algorithms":[235],"TSTC.":[237],"According":[238],"extensive":[240],"experiments,":[241],"achieves":[243],"1.24":[244],"x":[245,247,252],"~7.69":[246],"speedup":[248],"3.68":[250],"x~4.17":[251],"than":[255],"SOTA":[261],"customized":[262],"accelerator.":[263]},"counts_by_year":[{"year":2026,"cited_by_count":2},{"year":2025,"cited_by_count":3},{"year":2024,"cited_by_count":2}],"updated_date":"2026-04-02T15:55:50.835912","created_date":"2025-10-10T00:00:00"}