{"id":"https://openalex.org/W4411485108","doi":"https://doi.org/10.1109/tvlsi.2025.3577626","title":"Accelerating Unstructured Sparse DNNs via Multilevel Partial Sum Reduction and PE Array-Level Load Balancing","display_name":"Accelerating Unstructured Sparse DNNs via Multilevel Partial Sum Reduction and PE Array-Level Load Balancing","publication_year":2025,"publication_date":"2025-06-20","ids":{"openalex":"https://openalex.org/W4411485108","doi":"https://doi.org/10.1109/tvlsi.2025.3577626"},"language":"en","primary_location":{"id":"doi:10.1109/tvlsi.2025.3577626","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tvlsi.2025.3577626","pdf_url":null,"source":{"id":"https://openalex.org/S37538908","display_name":"IEEE Transactions on Very Large Scale Integration (VLSI) Systems","issn_l":"1063-8210","issn":["1063-8210","1557-9999"],"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 Very Large Scale Integration (VLSI) 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/A5118513586","display_name":"Chendong Xia","orcid":null},"institutions":[{"id":"https://openalex.org/I4210119392","display_name":"Institute of Microelectronics","ror":"https://ror.org/02s6gs133","country_code":"CN","type":"facility","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210119392"]},{"id":"https://openalex.org/I19820366","display_name":"Chinese Academy of Sciences","ror":"https://ror.org/034t30j35","country_code":"CN","type":"funder","lineage":["https://openalex.org/I19820366"]}],"countries":["CN"],"is_corresponding":true,"raw_author_name":"Chendong Xia","raw_affiliation_strings":["Institute of Microelectronics of the Chinese Academy of Sciences, Beijing, China"],"affiliations":[{"raw_affiliation_string":"Institute of Microelectronics of the Chinese Academy of Sciences, Beijing, China","institution_ids":["https://openalex.org/I4210119392","https://openalex.org/I19820366"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5103824017","display_name":"Qiang Li","orcid":"https://orcid.org/0000-0001-7774-6087"},"institutions":[{"id":"https://openalex.org/I19820366","display_name":"Chinese Academy of Sciences","ror":"https://ror.org/034t30j35","country_code":"CN","type":"funder","lineage":["https://openalex.org/I19820366"]},{"id":"https://openalex.org/I4210119392","display_name":"Institute of Microelectronics","ror":"https://ror.org/02s6gs133","country_code":"CN","type":"facility","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210119392"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Qiang Li","raw_affiliation_strings":["Institute of Microelectronics of the Chinese Academy of Sciences, Beijing, China"],"affiliations":[{"raw_affiliation_string":"Institute of Microelectronics of the Chinese Academy of Sciences, Beijing, China","institution_ids":["https://openalex.org/I4210119392","https://openalex.org/I19820366"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5107834173","display_name":"Zhi Li","orcid":null},"institutions":[{"id":"https://openalex.org/I4210119392","display_name":"Institute of Microelectronics","ror":"https://ror.org/02s6gs133","country_code":"CN","type":"facility","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210119392"]},{"id":"https://openalex.org/I19820366","display_name":"Chinese Academy of Sciences","ror":"https://ror.org/034t30j35","country_code":"CN","type":"funder","lineage":["https://openalex.org/I19820366"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zhi Li","raw_affiliation_strings":["Institute of Microelectronics of the Chinese Academy of Sciences, Beijing, China"],"affiliations":[{"raw_affiliation_string":"Institute of Microelectronics of the Chinese Academy of Sciences, Beijing, China","institution_ids":["https://openalex.org/I4210119392","https://openalex.org/I19820366"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100451305","display_name":"Bing Li","orcid":"https://orcid.org/0000-0003-0732-2267"},"institutions":[{"id":"https://openalex.org/I19820366","display_name":"Chinese Academy of Sciences","ror":"https://ror.org/034t30j35","country_code":"CN","type":"funder","lineage":["https://openalex.org/I19820366"]},{"id":"https://openalex.org/I4210119392","display_name":"Institute of Microelectronics","ror":"https://ror.org/02s6gs133","country_code":"CN","type":"facility","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210119392"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Bing Li","raw_affiliation_strings":["Institute of Microelectronics of the Chinese Academy of Sciences, Beijing, China"],"affiliations":[{"raw_affiliation_string":"Institute of Microelectronics of the Chinese Academy of Sciences, Beijing, China","institution_ids":["https://openalex.org/I4210119392","https://openalex.org/I19820366"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101267382","display_name":"Huidong Zhao","orcid":null},"institutions":[{"id":"https://openalex.org/I19820366","display_name":"Chinese Academy of Sciences","ror":"https://ror.org/034t30j35","country_code":"CN","type":"funder","lineage":["https://openalex.org/I19820366"]},{"id":"https://openalex.org/I4210119392","display_name":"Institute of Microelectronics","ror":"https://ror.org/02s6gs133","country_code":"CN","type":"facility","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210119392"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Huidong Zhao","raw_affiliation_strings":["Institute of Microelectronics of the Chinese Academy of Sciences, Beijing, China"],"affiliations":[{"raw_affiliation_string":"Institute of Microelectronics of the Chinese Academy of Sciences, Beijing, China","institution_ids":["https://openalex.org/I4210119392","https://openalex.org/I19820366"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5026870474","display_name":"Shushan Qiao","orcid":"https://orcid.org/0000-0002-9102-2111"},"institutions":[{"id":"https://openalex.org/I19820366","display_name":"Chinese Academy of Sciences","ror":"https://ror.org/034t30j35","country_code":"CN","type":"funder","lineage":["https://openalex.org/I19820366"]},{"id":"https://openalex.org/I4210119392","display_name":"Institute of Microelectronics","ror":"https://ror.org/02s6gs133","country_code":"CN","type":"facility","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210119392"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Shushan Qiao","raw_affiliation_strings":["Institute of Microelectronics of the Chinese Academy of Sciences, Beijing, China"],"affiliations":[{"raw_affiliation_string":"Institute of Microelectronics of the Chinese Academy of Sciences, Beijing, China","institution_ids":["https://openalex.org/I4210119392","https://openalex.org/I19820366"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":6,"corresponding_author_ids":["https://openalex.org/A5118513586"],"corresponding_institution_ids":["https://openalex.org/I19820366","https://openalex.org/I4210119392"],"apc_list":null,"apc_paid":null,"fwci":0.7467,"has_fulltext":false,"cited_by_count":1,"citation_normalized_percentile":{"value":0.73830465,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":91,"max":95},"biblio":{"volume":"33","issue":"8","first_page":"2329","last_page":"2333"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10187","display_name":"Radio Frequency Integrated Circuit Design","score":0.9771999716758728,"subfield":{"id":"https://openalex.org/subfields/2208","display_name":"Electrical and Electronic Engineering"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T10187","display_name":"Radio Frequency Integrated Circuit Design","score":0.9771999716758728,"subfield":{"id":"https://openalex.org/subfields/2208","display_name":"Electrical and Electronic Engineering"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10069","display_name":"Antenna Design and Analysis","score":0.9751999974250793,"subfield":{"id":"https://openalex.org/subfields/2202","display_name":"Aerospace Engineering"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T13182","display_name":"Quantum-Dot Cellular Automata","score":0.9509999752044678,"subfield":{"id":"https://openalex.org/subfields/1703","display_name":"Computational Theory and Mathematics"},"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/reduction","display_name":"Reduction (mathematics)","score":0.672990620136261},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.46453458070755005},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.29195478558540344},{"id":"https://openalex.org/keywords/geometry","display_name":"Geometry","score":0.04866376519203186}],"concepts":[{"id":"https://openalex.org/C111335779","wikidata":"https://www.wikidata.org/wiki/Q3454686","display_name":"Reduction (mathematics)","level":2,"score":0.672990620136261},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.46453458070755005},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.29195478558540344},{"id":"https://openalex.org/C2524010","wikidata":"https://www.wikidata.org/wiki/Q8087","display_name":"Geometry","level":1,"score":0.04866376519203186}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/tvlsi.2025.3577626","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tvlsi.2025.3577626","pdf_url":null,"source":{"id":"https://openalex.org/S37538908","display_name":"IEEE Transactions on Very Large Scale Integration (VLSI) Systems","issn_l":"1063-8210","issn":["1063-8210","1557-9999"],"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 Very Large Scale Integration (VLSI) Systems","raw_type":"journal-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":20,"referenced_works":["https://openalex.org/W2117539524","https://openalex.org/W2194775991","https://openalex.org/W2767737961","https://openalex.org/W2987129023","https://openalex.org/W3011962897","https://openalex.org/W3114479342","https://openalex.org/W3164217046","https://openalex.org/W4236868170","https://openalex.org/W4285507310","https://openalex.org/W4360606209","https://openalex.org/W4376455279","https://openalex.org/W4390097310","https://openalex.org/W4396605255","https://openalex.org/W4402349168","https://openalex.org/W4403445967","https://openalex.org/W6637373629","https://openalex.org/W6684191040","https://openalex.org/W6788135285","https://openalex.org/W6797478244","https://openalex.org/W6842795023"],"related_works":["https://openalex.org/W4391375266","https://openalex.org/W2899084033","https://openalex.org/W2748952813","https://openalex.org/W2390279801","https://openalex.org/W4391913857","https://openalex.org/W2358668433","https://openalex.org/W4396701345","https://openalex.org/W2376932109","https://openalex.org/W2001405890","https://openalex.org/W4396696052"],"abstract_inverted_index":{"Unstructured":[0],"pruning":[1],"introduces":[2],"significant":[3],"sparsity":[4,120],"in":[5,123],"deep":[6],"neural":[7],"networks":[8],"(DNNs),":[9],"enhancing":[10],"accelerator":[11,49,126],"hardware":[12],"efficiency.":[13],"However,":[14],"three":[15,54],"critical":[16],"challenges":[17],"constrain":[18],"performance":[19],"gains:":[20],"1)":[21],"complex":[22],"fetching":[23,72],"logic":[24],"for":[25],"nonzero":[26],"(NZ)":[27],"data":[28,68],"pairs;":[29],"2)":[30],"load":[31,93,100,111],"imbalance":[32],"across":[33,118],"processing":[34],"elements":[35],"(PEs);":[36],"and":[37,86,99,107,136,148],"3)":[38],"PE":[39,89,105,109,116],"stalls":[40,90],"from":[41],"write-back":[42,84],"contention.":[43],"This":[44],"brief":[45],"proposes":[46],"an":[47],"energy-efficient":[48],"addressing":[50],"these":[51],"inefficiencies":[52],"through":[53],"innovations.":[55],"First,":[56],"we":[57],"propose":[58],"a":[59,76,96],"Cartesian-product":[60],"output-row-stationary":[61],"(CPORS)":[62],"dataflow":[63],"that":[64],"inherently":[65],"matches":[66],"NZ":[67],"pairs":[69],"by":[70],"sequentially":[71],"compressed":[73],"data.":[74],"Second,":[75],"multilevel":[77],"partial":[78],"sum":[79],"reduction":[80],"(MLPR)":[81],"strategy":[82],"minimizes":[83],"traffic":[85],"converts":[87],"random":[88],"into":[91],"manageable":[92],"imbalance.":[94],"Third,":[95],"kernel":[97],"sorting":[98],"scheduling":[101],"(KSLS)":[102],"mechanism":[103],"resolves":[104],"idle/stall":[106],"achieves":[108,149],"array-level":[110],"balancing,":[112],"attaining":[113],"76.6%":[114],"average":[115],"utilization":[117],"all":[119],"levels.":[121],"Implemented":[122],"22-nmCMOS,":[124],"the":[125],"delivers":[127],"<inline-formula":[128,137],"xmlns:mml=\"http://www.w3.org/1998/Math/MathML\"":[129,138],"xmlns:xlink=\"http://www.w3.org/1999/xlink\">":[130,139],"<tex-math":[131,140],"notation=\"LaTeX\">$1.85\\times":[132],"$</tex-math>":[133,142],"</inline-formula>":[134,143],"speedup":[135],"notation=\"LaTeX\">$1.4\\times":[141],"energy":[144,153],"efficiency":[145,154],"over":[146],"baseline":[147],"25.8":[150],"TOPS/W":[151],"peak":[152],"at":[155],"90%":[156],"sparsity.":[157]},"counts_by_year":[{"year":2025,"cited_by_count":1}],"updated_date":"2025-11-06T03:46:38.306776","created_date":"2025-10-10T00:00:00"}