{"id":"https://openalex.org/W4406137499","doi":"https://doi.org/10.1109/tcsi.2024.3524307","title":"A Reconfigurable Floating-Point Division and Square Root Architecture for High-Precision Softmax","display_name":"A Reconfigurable Floating-Point Division and Square Root Architecture for High-Precision Softmax","publication_year":2025,"publication_date":"2025-01-07","ids":{"openalex":"https://openalex.org/W4406137499","doi":"https://doi.org/10.1109/tcsi.2024.3524307"},"language":"en","primary_location":{"id":"doi:10.1109/tcsi.2024.3524307","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tcsi.2024.3524307","pdf_url":null,"source":{"id":"https://openalex.org/S116977442","display_name":"IEEE Transactions on Circuits and Systems I Regular Papers","issn_l":"1549-8328","issn":["1549-8328","1558-0806"],"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 Circuits and Systems I: Regular Papers","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/A5016540097","display_name":"Xiwei Fang","orcid":"https://orcid.org/0009-0001-7523-0543"},"institutions":[{"id":"https://openalex.org/I3045169105","display_name":"Southern University of Science and Technology","ror":"https://ror.org/049tv2d57","country_code":"CN","type":"education","lineage":["https://openalex.org/I3045169105"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xiwei Fang","raw_affiliation_strings":["School of Microelectronics, Southern University of Science and Technology, Shenzhen, China"],"raw_orcid":"https://orcid.org/0009-0001-7523-0543","affiliations":[{"raw_affiliation_string":"School of Microelectronics, Southern University of Science and Technology, Shenzhen, China","institution_ids":["https://openalex.org/I3045169105"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100360658","display_name":"Yuhan Wang","orcid":"https://orcid.org/0000-0002-9376-6288"},"institutions":[{"id":"https://openalex.org/I3045169105","display_name":"Southern University of Science and Technology","ror":"https://ror.org/049tv2d57","country_code":"CN","type":"education","lineage":["https://openalex.org/I3045169105"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yuhan Wang","raw_affiliation_strings":["School of Microelectronics, Southern University of Science and Technology, Shenzhen, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Microelectronics, Southern University of Science and Technology, Shenzhen, China","institution_ids":["https://openalex.org/I3045169105"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100333431","display_name":"Lei Chen","orcid":"https://orcid.org/0000-0001-9936-3319"},"institutions":[{"id":"https://openalex.org/I3045169105","display_name":"Southern University of Science and Technology","ror":"https://ror.org/049tv2d57","country_code":"CN","type":"education","lineage":["https://openalex.org/I3045169105"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Lei Chen","raw_affiliation_strings":["School of Microelectronics, Southern University of Science and Technology, Shenzhen, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Microelectronics, Southern University of Science and Technology, Shenzhen, China","institution_ids":["https://openalex.org/I3045169105"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5069123107","display_name":"Fengwei An","orcid":"https://orcid.org/0000-0002-7554-7938"},"institutions":[{"id":"https://openalex.org/I3045169105","display_name":"Southern University of Science and Technology","ror":"https://ror.org/049tv2d57","country_code":"CN","type":"education","lineage":["https://openalex.org/I3045169105"]},{"id":"https://openalex.org/I4210136793","display_name":"Peng Cheng Laboratory","ror":"https://ror.org/03qdqbt06","country_code":"CN","type":"facility","lineage":["https://openalex.org/I4210136793"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Fengwei An","raw_affiliation_strings":["School of Microelectronics, Southern University of Science and Technology, Shenzhen, China","PengCheng Laboratory, Shenzhen, China"],"raw_orcid":"https://orcid.org/0000-0002-7554-7938","affiliations":[{"raw_affiliation_string":"School of Microelectronics, Southern University of Science and Technology, Shenzhen, China","institution_ids":["https://openalex.org/I3045169105"]},{"raw_affiliation_string":"PengCheng Laboratory, Shenzhen, China","institution_ids":["https://openalex.org/I4210136793"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":1.7708,"has_fulltext":false,"cited_by_count":2,"citation_normalized_percentile":{"value":0.83839057,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":91,"max":97},"biblio":{"volume":"72","issue":"9","first_page":"4737","last_page":"4750"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11697","display_name":"Numerical Methods and Algorithms","score":0.9997000098228455,"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"}},"topics":[{"id":"https://openalex.org/T11697","display_name":"Numerical Methods and Algorithms","score":0.9997000098228455,"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"}},{"id":"https://openalex.org/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.9987999796867371,"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/T10363","display_name":"Low-power high-performance VLSI design","score":0.9955999851226807,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/softmax-function","display_name":"Softmax function","score":0.8464987874031067},{"id":"https://openalex.org/keywords/pipeline","display_name":"Pipeline (software)","score":0.5819475054740906},{"id":"https://openalex.org/keywords/division","display_name":"Division (mathematics)","score":0.5740373134613037},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.5677849650382996},{"id":"https://openalex.org/keywords/reduction","display_name":"Reduction (mathematics)","score":0.5532379746437073},{"id":"https://openalex.org/keywords/floating-point","display_name":"Floating point","score":0.5304861664772034},{"id":"https://openalex.org/keywords/subtractor","display_name":"Subtractor","score":0.48412880301475525},{"id":"https://openalex.org/keywords/square-root","display_name":"Square root","score":0.4591066837310791},{"id":"https://openalex.org/keywords/cmos","display_name":"CMOS","score":0.43328163027763367},{"id":"https://openalex.org/keywords/embedded-system","display_name":"Embedded system","score":0.3640698492527008},{"id":"https://openalex.org/keywords/computer-hardware","display_name":"Computer hardware","score":0.34793058037757874},{"id":"https://openalex.org/keywords/electronic-engineering","display_name":"Electronic engineering","score":0.2883201539516449},{"id":"https://openalex.org/keywords/engineering","display_name":"Engineering","score":0.24975821375846863},{"id":"https://openalex.org/keywords/adder","display_name":"Adder","score":0.22365546226501465},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.21046018600463867},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.15766313672065735},{"id":"https://openalex.org/keywords/arithmetic","display_name":"Arithmetic","score":0.15405547618865967},{"id":"https://openalex.org/keywords/deep-learning","display_name":"Deep learning","score":0.1444040834903717},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.12233945727348328}],"concepts":[{"id":"https://openalex.org/C188441871","wikidata":"https://www.wikidata.org/wiki/Q7554146","display_name":"Softmax function","level":3,"score":0.8464987874031067},{"id":"https://openalex.org/C43521106","wikidata":"https://www.wikidata.org/wiki/Q2165493","display_name":"Pipeline (software)","level":2,"score":0.5819475054740906},{"id":"https://openalex.org/C60798267","wikidata":"https://www.wikidata.org/wiki/Q1226939","display_name":"Division (mathematics)","level":2,"score":0.5740373134613037},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5677849650382996},{"id":"https://openalex.org/C111335779","wikidata":"https://www.wikidata.org/wiki/Q3454686","display_name":"Reduction (mathematics)","level":2,"score":0.5532379746437073},{"id":"https://openalex.org/C84211073","wikidata":"https://www.wikidata.org/wiki/Q117879","display_name":"Floating point","level":2,"score":0.5304861664772034},{"id":"https://openalex.org/C187805909","wikidata":"https://www.wikidata.org/wiki/Q1142401","display_name":"Subtractor","level":4,"score":0.48412880301475525},{"id":"https://openalex.org/C11577676","wikidata":"https://www.wikidata.org/wiki/Q134237","display_name":"Square root","level":2,"score":0.4591066837310791},{"id":"https://openalex.org/C46362747","wikidata":"https://www.wikidata.org/wiki/Q173431","display_name":"CMOS","level":2,"score":0.43328163027763367},{"id":"https://openalex.org/C149635348","wikidata":"https://www.wikidata.org/wiki/Q193040","display_name":"Embedded system","level":1,"score":0.3640698492527008},{"id":"https://openalex.org/C9390403","wikidata":"https://www.wikidata.org/wiki/Q3966","display_name":"Computer hardware","level":1,"score":0.34793058037757874},{"id":"https://openalex.org/C24326235","wikidata":"https://www.wikidata.org/wiki/Q126095","display_name":"Electronic engineering","level":1,"score":0.2883201539516449},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.24975821375846863},{"id":"https://openalex.org/C164620267","wikidata":"https://www.wikidata.org/wiki/Q376953","display_name":"Adder","level":3,"score":0.22365546226501465},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.21046018600463867},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.15766313672065735},{"id":"https://openalex.org/C94375191","wikidata":"https://www.wikidata.org/wiki/Q11205","display_name":"Arithmetic","level":1,"score":0.15405547618865967},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.1444040834903717},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.12233945727348328},{"id":"https://openalex.org/C2524010","wikidata":"https://www.wikidata.org/wiki/Q8087","display_name":"Geometry","level":1,"score":0.0},{"id":"https://openalex.org/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/tcsi.2024.3524307","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tcsi.2024.3524307","pdf_url":null,"source":{"id":"https://openalex.org/S116977442","display_name":"IEEE Transactions on Circuits and Systems I Regular Papers","issn_l":"1549-8328","issn":["1549-8328","1558-0806"],"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 Circuits and Systems I: Regular Papers","raw_type":"journal-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G5128666534","display_name":null,"funder_award_id":"KQTD20200820113051096","funder_id":"https://openalex.org/F4320329836","funder_display_name":"Chuzhou Science and Technology Program"},{"id":"https://openalex.org/G851524468","display_name":null,"funder_award_id":"KJZD20230923113300002","funder_id":"https://openalex.org/F4320329836","funder_display_name":"Chuzhou Science and Technology Program"}],"funders":[{"id":"https://openalex.org/F4320329836","display_name":"Chuzhou Science and Technology Program","ror":null}],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":39,"referenced_works":["https://openalex.org/W1594525122","https://openalex.org/W1829595298","https://openalex.org/W1969048300","https://openalex.org/W2035378788","https://openalex.org/W2038017522","https://openalex.org/W2049305463","https://openalex.org/W2092427782","https://openalex.org/W2146659278","https://openalex.org/W2153039279","https://openalex.org/W2204310803","https://openalex.org/W2329007329","https://openalex.org/W2509774076","https://openalex.org/W2542333493","https://openalex.org/W2608093348","https://openalex.org/W2807198203","https://openalex.org/W2898915279","https://openalex.org/W2901085735","https://openalex.org/W2914247084","https://openalex.org/W2981278682","https://openalex.org/W2985363614","https://openalex.org/W3034075433","https://openalex.org/W3043532693","https://openalex.org/W3135047506","https://openalex.org/W3170233084","https://openalex.org/W3197218354","https://openalex.org/W3198509687","https://openalex.org/W4213019189","https://openalex.org/W4213100109","https://openalex.org/W4293768835","https://openalex.org/W4378364815","https://openalex.org/W4385625012","https://openalex.org/W4387385755","https://openalex.org/W4390873361","https://openalex.org/W4394627484","https://openalex.org/W4399151624","https://openalex.org/W4401210315","https://openalex.org/W4401567936","https://openalex.org/W4401935059","https://openalex.org/W4402159576"],"related_works":["https://openalex.org/W1981353157","https://openalex.org/W2048450681","https://openalex.org/W4238090084","https://openalex.org/W1944073140","https://openalex.org/W2153347456","https://openalex.org/W2070160710","https://openalex.org/W3200055559","https://openalex.org/W1969066812","https://openalex.org/W2592049182","https://openalex.org/W162485434"],"abstract_inverted_index":{"With":[0,215],"the":[1,7,21,80,89,106,109,122,125,151,158,162,169,180,220,241,244],"advancement":[2],"of":[3,20,164],"deep":[4],"learning":[5],"models,":[6],"Softmax":[8,22],"function":[9,23],"with":[10],"self-attention":[11],"has":[12],"become":[13],"pervasive":[14],"in":[15,168],"everyday":[16],"applications.":[17,148],"As":[18],"components":[19],"and":[24,29,44,61,71,87,92,154,188,201,206,227,258],"its":[25,34],"inputs,":[26],"both":[27],"division":[28,60,110,195,234,257],"square":[30,62,116,213,262],"root":[31,63,117],"operations":[32,40],"impact":[33],"accuracy.":[35],"However,":[36],"these":[37,52],"two":[38],"non-linear":[39],"bring":[41],"significant":[42],"area":[43,191,209,229],"power":[45,186,204,223],"consumption":[46,70,224],"for":[47,74,146,193,211,232,255,260],"hardware":[48,96,248],"implementation.":[49],"To":[50],"address":[51],"challenges,":[53],"this":[54],"paper":[55],"proposes":[56],"a":[57,115,130,138,184,189,207],"reconfigurable":[58,132],"floating-point":[59,194,212,233,256,261],"(FDSR)":[64],"architecture":[65,142],"that":[66,179],"achieves":[67,183],"low":[68],"resource":[69,97,249],"high":[72,101],"accuracy":[73,102],"general-purpose":[75],"computation.":[76],"The":[77],"FDSR":[78,159,221,245],"enhances":[79],"traditional":[81],"non-restoring":[82,170],"algorithm":[83],"by":[84,119,225,230],"using":[85,174],"shift-registers":[86],"optimizing":[88],"leading-one":[90],"detection":[91],"shift":[93],"operations,":[94],"reducing":[95],"usage":[98],"while":[99],"maintaining":[100],"(0.5":[103],"ULP).":[104],"In":[105],"mantissa":[107],"calculation,":[108],"can":[111],"be":[112],"converted":[113],"to":[114,124,143,197,236],"operation":[118],"simply":[120],"switching":[121],"input":[123],"subtractor":[126],"through":[127],"multiplexers.":[128],"Additionally,":[129],"triple-mode":[131],"iteration":[133,166],"unit":[134],"is":[135],"introduced,":[136],"featuring":[137],"multi-layer":[139],"variable":[140],"pipeline":[141,152],"improve":[144],"adaptability":[145],"different":[147],"By":[149],"redesigning":[150],"depth":[153],"reusing":[155],"logical":[156],"units,":[157],"effectively":[160],"addresses":[161],"issue":[163],"lengthy":[165],"cycles":[167],"method.":[171],"Implementation":[172],"results":[173],"40nm":[175],"CMOS":[176,218],"technology":[177],"demonstrate":[178],"proposed":[181],"design":[182],"76.49%":[185],"reduction":[187,192,205,210,254],"14.69%":[190],"compared":[196,235],"Synopsys":[198,237],"Design":[199,238],"Ware":[200],"an":[202,252],"88.05%":[203],"90.57%":[208],"root.":[214],"28":[216],"nm":[217],"technology,":[219],"reduces":[222,228,247],"91.55%":[226],"64.39%":[231],"Ware.":[239],"On":[240],"FPGA":[242],"platform,":[243],"significantly":[246],"consumption,":[250],"achieving":[251],"85.23%":[253],"87.81%":[259],"root,":[263],"outperforming":[264],"state-of-the-art":[265],"designs.":[266]},"counts_by_year":[{"year":2026,"cited_by_count":1},{"year":2025,"cited_by_count":1}],"updated_date":"2025-11-06T03:46:38.306776","created_date":"2025-10-10T00:00:00"}