{"id":"https://openalex.org/W7136134282","doi":"https://doi.org/10.1109/jiot.2026.3671312","title":"RISC-V Vectorized Softmax Acceleration for IoT Edge Inference Systems","display_name":"RISC-V Vectorized Softmax Acceleration for IoT Edge Inference Systems","publication_year":2026,"publication_date":"2026-03-16","ids":{"openalex":"https://openalex.org/W7136134282","doi":"https://doi.org/10.1109/jiot.2026.3671312"},"language":null,"primary_location":{"id":"doi:10.1109/jiot.2026.3671312","is_oa":false,"landing_page_url":"https://doi.org/10.1109/jiot.2026.3671312","pdf_url":null,"source":{"id":"https://openalex.org/S2480266640","display_name":"IEEE Internet of Things Journal","issn_l":"2327-4662","issn":["2327-4662","2372-2541"],"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 Internet of Things Journal","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/A5103237662","display_name":"Chin\u2010Feng Lai","orcid":"https://orcid.org/0000-0001-7138-0272"},"institutions":[{"id":"https://openalex.org/I91807558","display_name":"National Cheng Kung University","ror":"https://ror.org/01b8kcc49","country_code":"TW","type":"education","lineage":["https://openalex.org/I91807558"]}],"countries":["TW"],"is_corresponding":false,"raw_author_name":"Chin-Feng Lai","raw_affiliation_strings":["Department of Engineering Science, National Cheng Kung University, Tainan, Taiwan"],"raw_orcid":"https://orcid.org/0000-0001-7138-0272","affiliations":[{"raw_affiliation_string":"Department of Engineering Science, National Cheng Kung University, Tainan, Taiwan","institution_ids":["https://openalex.org/I91807558"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5126283691","display_name":"Po-Chih Liu","orcid":null},"institutions":[{"id":"https://openalex.org/I91807558","display_name":"National Cheng Kung University","ror":"https://ror.org/01b8kcc49","country_code":"TW","type":"education","lineage":["https://openalex.org/I91807558"]}],"countries":["TW"],"is_corresponding":false,"raw_author_name":"Po-Chih Liu","raw_affiliation_strings":["Department of Engineering Science, National Cheng Kung University, Tainan, Taiwan"],"raw_orcid":"https://orcid.org/0009-0000-5983-5908","affiliations":[{"raw_affiliation_string":"Department of Engineering Science, National Cheng Kung University, Tainan, Taiwan","institution_ids":["https://openalex.org/I91807558"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5129459785","display_name":"Shih-Hsin Huang","orcid":null},"institutions":[{"id":"https://openalex.org/I4210107330","display_name":"Powertech Technology (Taiwan)","ror":"https://ror.org/01c7pa495","country_code":"TW","type":"company","lineage":["https://openalex.org/I4210107330"]}],"countries":["TW"],"is_corresponding":false,"raw_author_name":"Shih-Hsin Huang","raw_affiliation_strings":["Andes Technology Corporation (Andes), Hsinchu, Taiwan"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Andes Technology Corporation (Andes), Hsinchu, Taiwan","institution_ids":["https://openalex.org/I4210107330"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5129572323","display_name":"Hsiao-Hwa Chen","orcid":null},"institutions":[{"id":"https://openalex.org/I91807558","display_name":"National Cheng Kung University","ror":"https://ror.org/01b8kcc49","country_code":"TW","type":"education","lineage":["https://openalex.org/I91807558"]}],"countries":["TW"],"is_corresponding":false,"raw_author_name":"Hsiao-Hwa Chen","raw_affiliation_strings":["Department of Engineering Science, National Cheng Kung University, Tainan, Taiwan"],"raw_orcid":"https://orcid.org/0000-0001-8473-0729","affiliations":[{"raw_affiliation_string":"Department of Engineering Science, National Cheng Kung University, Tainan, Taiwan","institution_ids":["https://openalex.org/I91807558"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.32805174,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":"13","issue":"12","first_page":"25890","last_page":"25901"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10904","display_name":"Embedded Systems Design Techniques","score":0.10159999877214432,"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/T10904","display_name":"Embedded Systems Design Techniques","score":0.10159999877214432,"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/T10273","display_name":"IoT and Edge/Fog Computing","score":0.09870000183582306,"subfield":{"id":"https://openalex.org/subfields/1705","display_name":"Computer Networks and Communications"},"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":0.04540000110864639,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/softmax-function","display_name":"Softmax function","score":0.7802000045776367},{"id":"https://openalex.org/keywords/acceleration","display_name":"Acceleration","score":0.7063999772071838},{"id":"https://openalex.org/keywords/enhanced-data-rates-for-gsm-evolution","display_name":"Enhanced Data Rates for GSM Evolution","score":0.5638999938964844},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.5181000232696533},{"id":"https://openalex.org/keywords/edge-computing","display_name":"Edge computing","score":0.4498000144958496},{"id":"https://openalex.org/keywords/internet-of-things","display_name":"Internet of Things","score":0.3968000113964081},{"id":"https://openalex.org/keywords/edge-device","display_name":"Edge device","score":0.39309999346733093}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8278999924659729},{"id":"https://openalex.org/C188441871","wikidata":"https://www.wikidata.org/wiki/Q7554146","display_name":"Softmax function","level":3,"score":0.7802000045776367},{"id":"https://openalex.org/C117896860","wikidata":"https://www.wikidata.org/wiki/Q11376","display_name":"Acceleration","level":2,"score":0.7063999772071838},{"id":"https://openalex.org/C162307627","wikidata":"https://www.wikidata.org/wiki/Q204833","display_name":"Enhanced Data Rates for GSM Evolution","level":2,"score":0.5638999938964844},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.5181000232696533},{"id":"https://openalex.org/C2778456923","wikidata":"https://www.wikidata.org/wiki/Q5337692","display_name":"Edge computing","level":3,"score":0.4498000144958496},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.39730000495910645},{"id":"https://openalex.org/C81860439","wikidata":"https://www.wikidata.org/wiki/Q251212","display_name":"Internet of Things","level":2,"score":0.3968000113964081},{"id":"https://openalex.org/C138236772","wikidata":"https://www.wikidata.org/wiki/Q25098575","display_name":"Edge device","level":3,"score":0.39309999346733093},{"id":"https://openalex.org/C193536780","wikidata":"https://www.wikidata.org/wiki/Q1513153","display_name":"Edge detection","level":4,"score":0.3370000123977661},{"id":"https://openalex.org/C79403827","wikidata":"https://www.wikidata.org/wiki/Q3988","display_name":"Real-time computing","level":1,"score":0.31450000405311584},{"id":"https://openalex.org/C67186912","wikidata":"https://www.wikidata.org/wiki/Q367664","display_name":"Data modeling","level":2,"score":0.29440000653266907},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.27900001406669617},{"id":"https://openalex.org/C106516650","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm design","level":2,"score":0.2786000072956085},{"id":"https://openalex.org/C2988105877","wikidata":"https://www.wikidata.org/wiki/Q1341685","display_name":"Inference system","level":5,"score":0.26190000772476196},{"id":"https://openalex.org/C104267543","wikidata":"https://www.wikidata.org/wiki/Q208163","display_name":"Signal processing","level":3,"score":0.2581000030040741},{"id":"https://openalex.org/C2777472644","wikidata":"https://www.wikidata.org/wiki/Q16968992","display_name":"Approximate inference","level":3,"score":0.2524999976158142}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/jiot.2026.3671312","is_oa":false,"landing_page_url":"https://doi.org/10.1109/jiot.2026.3671312","pdf_url":null,"source":{"id":"https://openalex.org/S2480266640","display_name":"IEEE Internet of Things Journal","issn_l":"2327-4662","issn":["2327-4662","2372-2541"],"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 Internet of Things Journal","raw_type":"journal-article"}],"best_oa_location":null,"sustainable_development_goals":[{"display_name":"Affordable and clean energy","score":0.6535976529121399,"id":"https://metadata.un.org/sdg/7"}],"awards":[{"id":"https://openalex.org/G1851290521","display_name":null,"funder_award_id":"114-2221-E-006-112","funder_id":"https://openalex.org/F4320331164","funder_display_name":"National Science and Technology Council"}],"funders":[{"id":"https://openalex.org/F4320331164","display_name":"National Science and Technology Council","ror":"https://ror.org/00wnb9798"}],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"In":[0,18],"Internet-of-Things":[1],"(IoT)":[2],"edge":[3,21,94,100,195],"inference":[4,196],"systems,":[5],"Softmax":[6,104,155],"is":[7,171],"a":[8,52,61,76,97,106,116,129,151],"core":[9],"component":[10],"in":[11,156,163,192],"multi-class":[12],"classification":[13],"heads":[14],"and":[15,38,58,115,145,158,166,177,183],"attention":[16],"mechanisms.":[17],"resource-constrained":[19],"IoT":[20,164,194],"devices,":[22],"conventional":[23],"numerically":[24,111],"stable":[25,112],"implementations":[26],"require":[27],"multiple":[28],"passes":[29],"over":[30,109,119],"the":[31,110,120],"data,":[32],"which":[33,87,188],"increases":[34],"memory":[35,66],"band-width":[36],"pressure":[37],"leads":[39],"to":[40,64,133],"hardly":[41],"predictable":[42],"variable":[43],"processing":[44],"latency.":[45],"This":[46],"paper":[47],"presents":[48],"Conditional":[49,102],"Online":[50,103],"Softmax,":[51,114],"control-flow-predictable":[53],"algorithm":[54],"that":[55],"fuses":[56],"reduction":[57],"normalization":[59],"into":[60],"single":[62],"pass":[63],"cut":[65],"traffic":[67],"while":[68],"preserving":[69],"numerical":[70],"stability.":[71],"We":[72],"pair":[73],"it":[74,150],"with":[75],"branch-free":[77],"vectorized":[78],"kernel":[79],"implemented":[80],"using":[81],"RISC-V":[82,99],"Vector":[83],"Extension":[84],"(RVV)":[85],"intrinsics,":[86],"better":[88],"exploits":[89],"instruction-level":[90],"parallelism":[91],"on":[92],"low-core-count":[93],"processors.":[95],"On":[96],"standard":[98,178],"platform,":[101],"achieves":[105],"7%":[107],"speedup":[108,118,131],"safe":[113],"43%":[117],"original":[121],"online":[122],"Softmax.":[123],"Furthermore,":[124],"manual":[125],"RVV":[126],"vectorization":[127],"delivers":[128],"2:9\u00d7":[130],"compared":[132],"GNU":[134],"Compiler":[135],"Collection":[136],"(GCC)":[137],"auto-vectorization.":[138],"The":[139,169],"resulting":[140],"design":[141],"reduces":[142],"cache":[143],"misses":[144],"improves":[146],"latency":[147],"predictability,":[148],"making":[149],"drop-in":[152],"replacement":[153],"for":[154],"streaming":[157],"micro-batch":[159],"pipelines":[160],"commonly":[161],"found":[162],"gateways":[165],"sensor":[167],"nodes.":[168],"implementation":[170],"portable":[172],"across":[173],"vector":[174],"length":[175],"(VLEN)":[176],"element":[179],"width":[180],"(SEW)":[181],"settings":[182],"requires":[184],"no":[185],"specialized":[186],"accelerators,":[187],"simplifies":[189],"its":[190],"deployment":[191],"practical":[193],"systems.":[197]},"counts_by_year":[],"updated_date":"2026-06-11T06:19:23.411458","created_date":"2026-03-17T00:00:00"}
