{"id":"https://openalex.org/W4416799067","doi":"https://doi.org/10.1109/apsipaasc65261.2025.11249034","title":"NOCTUA: A High-Efficiency Reconfigurable NoC-Based Transformer Universal Accelerator","display_name":"NOCTUA: A High-Efficiency Reconfigurable NoC-Based Transformer Universal Accelerator","publication_year":2025,"publication_date":"2025-10-22","ids":{"openalex":"https://openalex.org/W4416799067","doi":"https://doi.org/10.1109/apsipaasc65261.2025.11249034"},"language":null,"primary_location":{"id":"doi:10.1109/apsipaasc65261.2025.11249034","is_oa":false,"landing_page_url":"https://doi.org/10.1109/apsipaasc65261.2025.11249034","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 Asia Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA ASC)","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/A5088517936","display_name":"Kun-Chih Chen","orcid":"https://orcid.org/0000-0002-8908-468X"},"institutions":[{"id":"https://openalex.org/I148366613","display_name":"National Yang Ming Chiao Tung University","ror":"https://ror.org/00se2k293","country_code":"TW","type":"education","lineage":["https://openalex.org/I148366613"]}],"countries":["TW"],"is_corresponding":false,"raw_author_name":"Kun-Chih Chen","raw_affiliation_strings":["Institute of Electronics, National Yang Ming Chiao Tung University,Hsinchu,Taiwan"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Institute of Electronics, National Yang Ming Chiao Tung University,Hsinchu,Taiwan","institution_ids":["https://openalex.org/I148366613"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Pin-Ching Shen","orcid":null},"institutions":[{"id":"https://openalex.org/I148366613","display_name":"National Yang Ming Chiao Tung University","ror":"https://ror.org/00se2k293","country_code":"TW","type":"education","lineage":["https://openalex.org/I148366613"]}],"countries":["TW"],"is_corresponding":false,"raw_author_name":"Pin-Ching Shen","raw_affiliation_strings":["Institute of Electronics, National Yang Ming Chiao Tung University,Hsinchu,Taiwan"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Institute of Electronics, National Yang Ming Chiao Tung University,Hsinchu,Taiwan","institution_ids":["https://openalex.org/I148366613"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5071430992","display_name":"Bo-Chun Chen","orcid":null},"institutions":[{"id":"https://openalex.org/I148366613","display_name":"National Yang Ming Chiao Tung University","ror":"https://ror.org/00se2k293","country_code":"TW","type":"education","lineage":["https://openalex.org/I148366613"]}],"countries":["TW"],"is_corresponding":false,"raw_author_name":"Bo-Chun Chen","raw_affiliation_strings":["Institute of Electronics, National Yang Ming Chiao Tung University,Hsinchu,Taiwan"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Institute of Electronics, National Yang Ming Chiao Tung University,Hsinchu,Taiwan","institution_ids":["https://openalex.org/I148366613"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I148366613"],"apc_list":null,"apc_paid":null,"fwci":1.7646,"has_fulltext":false,"cited_by_count":2,"citation_normalized_percentile":{"value":0.87826901,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":97,"max":98},"biblio":{"volume":null,"issue":null,"first_page":"1270","last_page":"1273"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.2583000063896179,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"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/T10036","display_name":"Advanced Neural Network Applications","score":0.2583000063896179,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"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":0.11050000041723251,"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/T11948","display_name":"Machine Learning in Materials Science","score":0.10530000180006027,"subfield":{"id":"https://openalex.org/subfields/2505","display_name":"Materials Chemistry"},"field":{"id":"https://openalex.org/fields/25","display_name":"Materials Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/scalability","display_name":"Scalability","score":0.6779000163078308},{"id":"https://openalex.org/keywords/transformer","display_name":"Transformer","score":0.590499997138977},{"id":"https://openalex.org/keywords/dataflow","display_name":"Dataflow","score":0.49880000948905945},{"id":"https://openalex.org/keywords/efficient-energy-use","display_name":"Efficient energy use","score":0.49729999899864197},{"id":"https://openalex.org/keywords/field-programmable-gate-array","display_name":"Field-programmable gate array","score":0.484499990940094}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6829000115394592},{"id":"https://openalex.org/C48044578","wikidata":"https://www.wikidata.org/wiki/Q727490","display_name":"Scalability","level":2,"score":0.6779000163078308},{"id":"https://openalex.org/C66322947","wikidata":"https://www.wikidata.org/wiki/Q11658","display_name":"Transformer","level":3,"score":0.590499997138977},{"id":"https://openalex.org/C96324660","wikidata":"https://www.wikidata.org/wiki/Q205446","display_name":"Dataflow","level":2,"score":0.49880000948905945},{"id":"https://openalex.org/C2742236","wikidata":"https://www.wikidata.org/wiki/Q924713","display_name":"Efficient energy use","level":2,"score":0.49729999899864197},{"id":"https://openalex.org/C42935608","wikidata":"https://www.wikidata.org/wiki/Q190411","display_name":"Field-programmable gate array","level":2,"score":0.484499990940094},{"id":"https://openalex.org/C118524514","wikidata":"https://www.wikidata.org/wiki/Q173212","display_name":"Computer architecture","level":1,"score":0.47429999709129333},{"id":"https://openalex.org/C149635348","wikidata":"https://www.wikidata.org/wiki/Q193040","display_name":"Embedded system","level":1,"score":0.4074999988079071},{"id":"https://openalex.org/C123745756","wikidata":"https://www.wikidata.org/wiki/Q1665949","display_name":"Interconnection","level":2,"score":0.3935000002384186},{"id":"https://openalex.org/C2780598303","wikidata":"https://www.wikidata.org/wiki/Q65921492","display_name":"Flexibility (engineering)","level":2,"score":0.36039999127388},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.3118000030517578},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.30869999527931213},{"id":"https://openalex.org/C9390403","wikidata":"https://www.wikidata.org/wiki/Q3966","display_name":"Computer hardware","level":1,"score":0.262800008058548},{"id":"https://openalex.org/C101468663","wikidata":"https://www.wikidata.org/wiki/Q1620158","display_name":"Modular design","level":2,"score":0.26179999113082886}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/apsipaasc65261.2025.11249034","is_oa":false,"landing_page_url":"https://doi.org/10.1109/apsipaasc65261.2025.11249034","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 Asia Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA ASC)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":11,"referenced_works":["https://openalex.org/W2160642395","https://openalex.org/W2604319603","https://openalex.org/W2606722458","https://openalex.org/W2896457183","https://openalex.org/W3082020764","https://openalex.org/W3085139254","https://openalex.org/W4280582299","https://openalex.org/W4379116023","https://openalex.org/W4384947633","https://openalex.org/W4385245566","https://openalex.org/W4386859272"],"related_works":[],"abstract_inverted_index":{"Transformer":[0,28,39,69,121],"models,":[1,154],"particularly":[2],"Large":[3],"Language":[4],"Models":[5],"(LLMs),":[6],"are":[7,64],"revolutionizing":[8],"numerous":[9],"fields":[10],"but":[11],"impose":[12],"significant":[13,135],"computational":[14],"demands,":[15],"necessitating":[16],"specialized":[17],"hardware":[18,103],"acceleration.":[19],"This":[20,86],"paper":[21],"introduces":[22],"NOCTUA,":[23],"a":[24,35,44,56,117,165],"novel":[25],"Network-On-Chip":[26],"(NoC)-based":[27],"Universal":[29,40],"Accelerator,":[30],"designed":[31,65],"and":[32,51,73,83,88,105,113,123,132,141,157,174],"presented":[33],"as":[34],"High-Efficiency":[36],"Reconfigurable":[37],"NoC-based":[38],"Accelerator.":[41],"NOCTUA":[42,98,131,163],"leverages":[43],"scalable":[45],"Network-on-Chip,":[46],"renowned":[47],"for":[48],"its":[49,102,134],"flexibility":[50],"high-bandwidth":[52],"communication,":[53],"to":[54,66,99,137,145],"interconnect":[55],"multitude":[57],"of":[58,120,130,167],"Processing":[59],"Elements":[60],"(PEs).":[61],"These":[62],"PEs":[63],"execute":[67],"complex":[68],"operations":[70],"both":[71,109],"independently":[72],"in":[74,178],"coordinated":[75],"concert":[76],"with":[77,150,181],"other":[78],"PEs,":[79],"enabling":[80],"fine-grained":[81],"parallelism":[82],"collaborative":[84],"computation.":[85],"distributed":[87],"reconfigurable":[89],"design,":[90],"underpinned":[91],"by":[92],"the":[93,127],"adaptable":[94],"NoC":[95],"backbone,":[96],"allows":[97],"dynamically":[100],"adapt":[101],"resources":[104],"dataflow":[106],"pathways,":[107],"achieving":[108],"high":[110],"operational":[111],"efficiency":[112,143,180],"universal":[114],"applicability":[115],"across":[116],"wide":[118],"spectrum":[119],"models":[122,177],"workloads.":[124],"We":[125],"present":[126],"architectural":[128],"details":[129],"demonstrate":[133],"potential":[136],"enhance":[138],"processing":[139],"throughput":[140],"energy":[142],"compared":[144],"conventional":[146],"approaches.":[147],"Through":[148],"experiments":[149],"existing":[151],"attention-based":[152],"NLP":[153],"including":[155],"BERT":[156],"GPT-2":[158],"on":[159],"various":[160],"language":[161],"tasks,":[162],"achieves":[164],"performance":[166],"<tex":[168,184],"xmlns:mml=\"http://www.w3.org/1998/Math/MathML\"":[169,185],"xmlns:xlink=\"http://www.w3.org/1999/xlink\">$\\mathbf{1.":[170],"0":[171],"5}$</tex>":[172],"TOPS/W":[173],"outperforms":[175],"these":[176],"area":[179],"0.71":[182],"TOPS":[183],"xmlns:xlink=\"http://www.w3.org/1999/xlink\">$/":[186],"\\text{mm}^{2}$</tex>.":[187]},"counts_by_year":[{"year":2026,"cited_by_count":2}],"updated_date":"2026-06-26T08:34:08.712188","created_date":"2025-11-28T00:00:00"}