{"id":"https://openalex.org/W4412397574","doi":"https://doi.org/10.1177/10943420251355115","title":"Characterization of quantized inference with transformer encoders on low power CPUs","display_name":"Characterization of quantized inference with transformer encoders on low power CPUs","publication_year":2025,"publication_date":"2025-07-13","ids":{"openalex":"https://openalex.org/W4412397574","doi":"https://doi.org/10.1177/10943420251355115"},"language":"en","primary_location":{"id":"doi:10.1177/10943420251355115","is_oa":false,"landing_page_url":"https://doi.org/10.1177/10943420251355115","pdf_url":null,"source":{"id":"https://openalex.org/S60606485","display_name":"The International Journal of High Performance Computing Applications","issn_l":"1094-3420","issn":["1094-3420","1741-2846"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310320017","host_organization_name":"SAGE Publishing","host_organization_lineage":["https://openalex.org/P4310320017"],"host_organization_lineage_names":["SAGE Publishing"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"The International Journal of High Performance Computing Applications","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/A5018719028","display_name":"H\u00e8ctor Mart\u00ednez","orcid":"https://orcid.org/0000-0001-5891-4479"},"institutions":[{"id":"https://openalex.org/I53110688","display_name":"University of C\u00f3rdoba","ror":"https://ror.org/05yc77b46","country_code":"ES","type":"education","lineage":["https://openalex.org/I53110688"]}],"countries":["ES"],"is_corresponding":false,"raw_author_name":"H\u00e9ctor Mart\u00ednez","raw_affiliation_strings":["Universidad de C\u00f3rdoba","Departamento de Ingenier\u00eda Electr\u00f3nica y de Computadores, Universidad de C\u00f3rdoba, C\u00f3rdoba, Spain"],"affiliations":[{"raw_affiliation_string":"Universidad de C\u00f3rdoba","institution_ids":[]},{"raw_affiliation_string":"Departamento de Ingenier\u00eda Electr\u00f3nica y de Computadores, Universidad de C\u00f3rdoba, C\u00f3rdoba, Spain","institution_ids":["https://openalex.org/I53110688"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5005202822","display_name":"Sandra Catal\u00e1n","orcid":"https://orcid.org/0000-0002-9321-2728"},"institutions":[{"id":"https://openalex.org/I10902133","display_name":"Universitat Jaume I","ror":"https://ror.org/02ws1xc11","country_code":"ES","type":"education","lineage":["https://openalex.org/I10902133"]}],"countries":["ES"],"is_corresponding":true,"raw_author_name":"Sandra Catal\u00e1n","raw_affiliation_strings":["Universitat Jaume I","Departament d'Enginyeria i Ci\u00e8ncia de Computadors, Universitat Jaume I, Castell\u00f3 de la Plana, Spain"],"affiliations":[{"raw_affiliation_string":"Universitat Jaume I","institution_ids":["https://openalex.org/I10902133"]},{"raw_affiliation_string":"Departament d'Enginyeria i Ci\u00e8ncia de Computadors, Universitat Jaume I, Castell\u00f3 de la Plana, Spain","institution_ids":["https://openalex.org/I10902133"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5033947083","display_name":"Adri\u00e1n Castell\u00f3","orcid":"https://orcid.org/0000-0002-8576-8451"},"institutions":[{"id":"https://openalex.org/I60053951","display_name":"Universitat Polit\u00e8cnica de Val\u00e8ncia","ror":"https://ror.org/01460j859","country_code":"ES","type":"education","lineage":["https://openalex.org/I60053951"]}],"countries":["ES"],"is_corresponding":false,"raw_author_name":"Adri\u00e1n Castell\u00f3","raw_affiliation_strings":["Universitat Polit\u00e8cnica de Val\u00e8ncia","Departamento de Inform\u00e1tica de Sistemas y Computadores, Universitat Polit\u00e8cnica de Val\u00e8ncia, Val\u00e8ncia, Spain"],"affiliations":[{"raw_affiliation_string":"Universitat Polit\u00e8cnica de Val\u00e8ncia","institution_ids":["https://openalex.org/I60053951"]},{"raw_affiliation_string":"Departamento de Inform\u00e1tica de Sistemas y Computadores, Universitat Polit\u00e8cnica de Val\u00e8ncia, Val\u00e8ncia, Spain","institution_ids":["https://openalex.org/I60053951"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5012806004","display_name":"Enrique S. Quintana\u2013Ort\u0301\u0131","orcid":"https://orcid.org/0000-0002-5454-165X"},"institutions":[{"id":"https://openalex.org/I60053951","display_name":"Universitat Polit\u00e8cnica de Val\u00e8ncia","ror":"https://ror.org/01460j859","country_code":"ES","type":"education","lineage":["https://openalex.org/I60053951"]}],"countries":["ES"],"is_corresponding":false,"raw_author_name":"Enrique S Quintana-Ort\u00ed","raw_affiliation_strings":["Universitat Polit\u00e8cnica de Val\u00e8ncia","Departamento de Inform\u00e1tica de Sistemas y Computadores, Universitat Polit\u00e8cnica de Val\u00e8ncia, Val\u00e8ncia, Spain"],"affiliations":[{"raw_affiliation_string":"Universitat Polit\u00e8cnica de Val\u00e8ncia","institution_ids":["https://openalex.org/I60053951"]},{"raw_affiliation_string":"Departamento de Inform\u00e1tica de Sistemas y Computadores, Universitat Polit\u00e8cnica de Val\u00e8ncia, Val\u00e8ncia, Spain","institution_ids":["https://openalex.org/I60053951"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":4,"corresponding_author_ids":["https://openalex.org/A5005202822"],"corresponding_institution_ids":["https://openalex.org/I10902133"],"apc_list":null,"apc_paid":null,"fwci":2.4849,"has_fulltext":false,"cited_by_count":1,"citation_normalized_percentile":{"value":0.90463821,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":96,"max":99},"biblio":{"volume":"39","issue":"6","first_page":"803","last_page":"821"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10320","display_name":"Neural Networks and Applications","score":0.9973000288009644,"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"}},"topics":[{"id":"https://openalex.org/T10320","display_name":"Neural Networks and Applications","score":0.9973000288009644,"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"}},{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.9923999905586243,"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/T12814","display_name":"Gaussian Processes and Bayesian Inference","score":0.9842000007629395,"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/inference","display_name":"Inference","score":0.6802269816398621},{"id":"https://openalex.org/keywords/transformer","display_name":"Transformer","score":0.6068732142448425},{"id":"https://openalex.org/keywords/encoder","display_name":"Encoder","score":0.6061998009681702},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.49008214473724365},{"id":"https://openalex.org/keywords/electrical-engineering","display_name":"Electrical engineering","score":0.23325681686401367},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.18345841765403748},{"id":"https://openalex.org/keywords/engineering","display_name":"Engineering","score":0.1568712294101715},{"id":"https://openalex.org/keywords/operating-system","display_name":"Operating system","score":0.11589601635932922},{"id":"https://openalex.org/keywords/voltage","display_name":"Voltage","score":0.05555126070976257}],"concepts":[{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.6802269816398621},{"id":"https://openalex.org/C66322947","wikidata":"https://www.wikidata.org/wiki/Q11658","display_name":"Transformer","level":3,"score":0.6068732142448425},{"id":"https://openalex.org/C118505674","wikidata":"https://www.wikidata.org/wiki/Q42586063","display_name":"Encoder","level":2,"score":0.6061998009681702},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.49008214473724365},{"id":"https://openalex.org/C119599485","wikidata":"https://www.wikidata.org/wiki/Q43035","display_name":"Electrical engineering","level":1,"score":0.23325681686401367},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.18345841765403748},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.1568712294101715},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.11589601635932922},{"id":"https://openalex.org/C165801399","wikidata":"https://www.wikidata.org/wiki/Q25428","display_name":"Voltage","level":2,"score":0.05555126070976257}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1177/10943420251355115","is_oa":false,"landing_page_url":"https://doi.org/10.1177/10943420251355115","pdf_url":null,"source":{"id":"https://openalex.org/S60606485","display_name":"The International Journal of High Performance Computing Applications","issn_l":"1094-3420","issn":["1094-3420","1741-2846"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310320017","host_organization_name":"SAGE Publishing","host_organization_lineage":["https://openalex.org/P4310320017"],"host_organization_lineage_names":["SAGE Publishing"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"The International Journal of High Performance Computing Applications","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":26,"referenced_works":["https://openalex.org/W2043275593","https://openalex.org/W2073061372","https://openalex.org/W2154865023","https://openalex.org/W2251939518","https://openalex.org/W2252007067","https://openalex.org/W2412152731","https://openalex.org/W2516525699","https://openalex.org/W2524428287","https://openalex.org/W2949953997","https://openalex.org/W3099506637","https://openalex.org/W3130954105","https://openalex.org/W3142264480","https://openalex.org/W3170233084","https://openalex.org/W3210601829","https://openalex.org/W4244254628","https://openalex.org/W4308090443","https://openalex.org/W4322718253","https://openalex.org/W4362515116","https://openalex.org/W4378697133","https://openalex.org/W4385245566","https://openalex.org/W4386826409","https://openalex.org/W4401861582","https://openalex.org/W4405390006","https://openalex.org/W6600228653","https://openalex.org/W6600445788","https://openalex.org/W6739901393"],"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/W4206178588","https://openalex.org/W3094491777","https://openalex.org/W3214715529","https://openalex.org/W4287635093"],"abstract_inverted_index":{"Transformers,":[0],"particularly":[1],"large":[2],"language":[3,11],"models":[4],"(LLMs),":[5],"are":[6],"revolutionizing":[7],"applications":[8],"in":[9,21,72],"natural":[10],"processing":[12],"and":[13,24,45,105,108,112,134],"computer":[14],"vision":[15],"but":[16],"at":[17],"a":[18,92],"high":[19],"cost":[20],"memory,":[22],"energy,":[23],"computational":[25],"resources.":[26],"Quantization":[27],"has":[28],"emerged":[29],"as":[30,68],"an":[31],"effective":[32],"compression":[33],"method":[34],"to":[35,48,128],"alleviate":[36],"these":[37],"demands,":[38],"reducing":[39],"the":[40,80,123],"bitwidth":[41],"of":[42,83,95,115,125],"model":[43],"data":[44],"arithmetic":[46],"precision":[47,100],"enable":[49],"efficient":[50,142],"inference":[51,60],"on":[52,58,64,89,103,131,145],"resource-constrained":[53],"devices.":[54,74],"This":[55],"paper":[56],"focuses":[57],"optimizing":[59],"with":[61,91],"transformer":[62,87,129],"encoders":[63,88,130],"low-power":[65],"general-purpose":[66],"CPUs,":[67,138],"those":[69],"often":[70],"found":[71],"edge":[73,146],"Our":[75],"key":[76],"contributions":[77],"include":[78],"exposing":[79],"critical":[81],"role":[82],"linear":[84],"layers":[85],"within":[86],"CPUs":[90],"limited":[93],"number":[94],"cores;":[96],"developing":[97],"mixed":[98],"integer":[99],"matrix":[101],"multiplication":[102],"ARM":[104],"RISC-V":[106],"CPUs;":[107],"evaluating":[109],"performance":[110],"impact":[111],"energy":[113],"savings":[114],"quantized":[116],"inference.":[117],"In":[118],"summary,":[119],"this":[120],"work":[121],"highlights":[122],"advantages":[124],"applying":[126],"quantization":[127],"current":[132],"single-core":[133],"multi-core":[135],"low":[136],"power":[137],"offering":[139],"insights":[140],"for":[141],"LLM":[143],"deployment":[144],"platforms.":[147]},"counts_by_year":[{"year":2026,"cited_by_count":1}],"updated_date":"2026-03-27T05:58:40.876381","created_date":"2025-10-10T00:00:00"}
