{"id":"https://openalex.org/W4404609150","doi":"https://doi.org/10.1109/tc.2024.3500369","title":"Mix-GEMM: Extending RISC-V CPUs for\u00a0Energy-Efficient Mixed-Precision DNN Inference\u00a0Using Binary\u00a0Segmentation","display_name":"Mix-GEMM: Extending RISC-V CPUs for\u00a0Energy-Efficient Mixed-Precision DNN Inference\u00a0Using Binary\u00a0Segmentation","publication_year":2024,"publication_date":"2024-11-21","ids":{"openalex":"https://openalex.org/W4404609150","doi":"https://doi.org/10.1109/tc.2024.3500369"},"language":"en","primary_location":{"id":"doi:10.1109/tc.2024.3500369","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tc.2024.3500369","pdf_url":null,"source":{"id":"https://openalex.org/S157670870","display_name":"IEEE Transactions on Computers","issn_l":"0018-9340","issn":["0018-9340","1557-9956","2326-3814"],"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 Computers","raw_type":"journal-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://upcommons.upc.edu/bitstreams/1cca3fd2-04ca-4d45-b44b-b7fdf16f2358/download","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5079867677","display_name":"Jordi Fornt","orcid":"https://orcid.org/0000-0002-2484-9688"},"institutions":[{"id":"https://openalex.org/I2799803557","display_name":"Barcelona Supercomputing Center","ror":"https://ror.org/05sd8tv96","country_code":"ES","type":"facility","lineage":["https://openalex.org/I2799803557","https://openalex.org/I9617848"]},{"id":"https://openalex.org/I9617848","display_name":"Universitat Polit\u00e8cnica de Catalunya","ror":"https://ror.org/03mb6wj31","country_code":"ES","type":"education","lineage":["https://openalex.org/I9617848"]}],"countries":["ES"],"is_corresponding":true,"raw_author_name":"Jordi Fornt","raw_affiliation_strings":["Barcelona Supercomputing Center (BSC), Barcelona, Spain","Barcelona Supercomputing Center (BSC)Universitat Polit`ecnica de Catalunya (UPC)"],"affiliations":[{"raw_affiliation_string":"Barcelona Supercomputing Center (BSC), Barcelona, Spain","institution_ids":["https://openalex.org/I9617848","https://openalex.org/I2799803557"]},{"raw_affiliation_string":"Barcelona Supercomputing Center (BSC)Universitat Polit`ecnica de Catalunya (UPC)","institution_ids":["https://openalex.org/I2799803557","https://openalex.org/I9617848"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5064165125","display_name":"Enrico Reggiani","orcid":"https://orcid.org/0000-0003-1385-7962"},"institutions":[{"id":"https://openalex.org/I2799803557","display_name":"Barcelona Supercomputing Center","ror":"https://ror.org/05sd8tv96","country_code":"ES","type":"facility","lineage":["https://openalex.org/I2799803557","https://openalex.org/I9617848"]},{"id":"https://openalex.org/I9617848","display_name":"Universitat Polit\u00e8cnica de Catalunya","ror":"https://ror.org/03mb6wj31","country_code":"ES","type":"education","lineage":["https://openalex.org/I9617848"]}],"countries":["ES"],"is_corresponding":false,"raw_author_name":"Enrico Reggiani","raw_affiliation_strings":["Barcelona Supercomputing Center (BSC), Barcelona, Spain"],"affiliations":[{"raw_affiliation_string":"Barcelona Supercomputing Center (BSC), Barcelona, Spain","institution_ids":["https://openalex.org/I9617848","https://openalex.org/I2799803557"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5000595364","display_name":"Pau Fontova-Must\u00e9","orcid":"https://orcid.org/0000-0003-1839-416X"},"institutions":[{"id":"https://openalex.org/I2799803557","display_name":"Barcelona Supercomputing Center","ror":"https://ror.org/05sd8tv96","country_code":"ES","type":"facility","lineage":["https://openalex.org/I2799803557","https://openalex.org/I9617848"]},{"id":"https://openalex.org/I9617848","display_name":"Universitat Polit\u00e8cnica de Catalunya","ror":"https://ror.org/03mb6wj31","country_code":"ES","type":"education","lineage":["https://openalex.org/I9617848"]}],"countries":["ES"],"is_corresponding":false,"raw_author_name":"Pau Fontova-Must\u00e9","raw_affiliation_strings":["Barcelona Supercomputing Center (BSC), Barcelona, Spain"],"affiliations":[{"raw_affiliation_string":"Barcelona Supercomputing Center (BSC), Barcelona, Spain","institution_ids":["https://openalex.org/I9617848","https://openalex.org/I2799803557"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5074423970","display_name":"Narc\u00eds Rodas","orcid":null},"institutions":[{"id":"https://openalex.org/I2799803557","display_name":"Barcelona Supercomputing Center","ror":"https://ror.org/05sd8tv96","country_code":"ES","type":"facility","lineage":["https://openalex.org/I2799803557","https://openalex.org/I9617848"]},{"id":"https://openalex.org/I9617848","display_name":"Universitat Polit\u00e8cnica de Catalunya","ror":"https://ror.org/03mb6wj31","country_code":"ES","type":"education","lineage":["https://openalex.org/I9617848"]}],"countries":["ES"],"is_corresponding":false,"raw_author_name":"Narc\u00eds Rodas","raw_affiliation_strings":["Barcelona Supercomputing Center (BSC), Barcelona, Spain"],"affiliations":[{"raw_affiliation_string":"Barcelona Supercomputing Center (BSC), Barcelona, Spain","institution_ids":["https://openalex.org/I9617848","https://openalex.org/I2799803557"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5076669165","display_name":"Alessandro Pappalardo","orcid":"https://orcid.org/0000-0001-9386-5510"},"institutions":[{"id":"https://openalex.org/I1311921367","display_name":"Advanced Micro Devices (Canada)","ror":"https://ror.org/02yh0k313","country_code":"CA","type":"company","lineage":["https://openalex.org/I1311921367","https://openalex.org/I4210137977"]},{"id":"https://openalex.org/I4210137977","display_name":"Advanced Micro Devices (United States)","ror":"https://ror.org/04kd6c783","country_code":"US","type":"company","lineage":["https://openalex.org/I4210137977"]}],"countries":["CA","US"],"is_corresponding":false,"raw_author_name":"Alessandro Pappalardo","raw_affiliation_strings":["Advanced Micro-Devices (AMD), Dublin, Ireland","Advanced Micro-Devices (AMD)"],"affiliations":[{"raw_affiliation_string":"Advanced Micro-Devices (AMD), Dublin, Ireland","institution_ids":["https://openalex.org/I4210137977","https://openalex.org/I1311921367"]},{"raw_affiliation_string":"Advanced Micro-Devices (AMD)","institution_ids":["https://openalex.org/I1311921367"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5075162875","display_name":"Osman \u00dcnsal","orcid":"https://orcid.org/0000-0002-0544-9697"},"institutions":[{"id":"https://openalex.org/I2799803557","display_name":"Barcelona Supercomputing Center","ror":"https://ror.org/05sd8tv96","country_code":"ES","type":"facility","lineage":["https://openalex.org/I2799803557","https://openalex.org/I9617848"]},{"id":"https://openalex.org/I9617848","display_name":"Universitat Polit\u00e8cnica de Catalunya","ror":"https://ror.org/03mb6wj31","country_code":"ES","type":"education","lineage":["https://openalex.org/I9617848"]}],"countries":["ES"],"is_corresponding":false,"raw_author_name":"Osman Sabri Unsal","raw_affiliation_strings":["Barcelona Supercomputing Center (BSC), Barcelona, Spain"],"affiliations":[{"raw_affiliation_string":"Barcelona Supercomputing Center (BSC), Barcelona, Spain","institution_ids":["https://openalex.org/I9617848","https://openalex.org/I2799803557"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5084390427","display_name":"Adri\u00e1n Cristal","orcid":"https://orcid.org/0000-0003-1277-9296"},"institutions":[{"id":"https://openalex.org/I2799803557","display_name":"Barcelona Supercomputing Center","ror":"https://ror.org/05sd8tv96","country_code":"ES","type":"facility","lineage":["https://openalex.org/I2799803557","https://openalex.org/I9617848"]},{"id":"https://openalex.org/I9617848","display_name":"Universitat Polit\u00e8cnica de Catalunya","ror":"https://ror.org/03mb6wj31","country_code":"ES","type":"education","lineage":["https://openalex.org/I9617848"]}],"countries":["ES"],"is_corresponding":false,"raw_author_name":"Adri\u00e1n Cristal Kestelman","raw_affiliation_strings":["Barcelona Supercomputing Center (BSC), Barcelona, Spain","Barcelona Supercomputing Center (BSC)Universitat Polit`ecnica de Catalunya (UPC)"],"affiliations":[{"raw_affiliation_string":"Barcelona Supercomputing Center (BSC), Barcelona, Spain","institution_ids":["https://openalex.org/I9617848","https://openalex.org/I2799803557"]},{"raw_affiliation_string":"Barcelona Supercomputing Center (BSC)Universitat Polit`ecnica de Catalunya (UPC)","institution_ids":["https://openalex.org/I2799803557","https://openalex.org/I9617848"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5038596298","display_name":"Josep Altet","orcid":"https://orcid.org/0000-0002-6939-6475"},"institutions":[{"id":"https://openalex.org/I9617848","display_name":"Universitat Polit\u00e8cnica de Catalunya","ror":"https://ror.org/03mb6wj31","country_code":"ES","type":"education","lineage":["https://openalex.org/I9617848"]}],"countries":["ES"],"is_corresponding":false,"raw_author_name":"Josep Altet","raw_affiliation_strings":["Universitat Polit&#x00E8;cnica de Catalunya (UPC), Barcelona, Spain"],"affiliations":[{"raw_affiliation_string":"Universitat Polit&#x00E8;cnica de Catalunya (UPC), Barcelona, Spain","institution_ids":["https://openalex.org/I9617848"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5028715959","display_name":"Francesc Moll","orcid":"https://orcid.org/0000-0002-1290-3253"},"institutions":[{"id":"https://openalex.org/I2799803557","display_name":"Barcelona Supercomputing Center","ror":"https://ror.org/05sd8tv96","country_code":"ES","type":"facility","lineage":["https://openalex.org/I2799803557","https://openalex.org/I9617848"]},{"id":"https://openalex.org/I9617848","display_name":"Universitat Polit\u00e8cnica de Catalunya","ror":"https://ror.org/03mb6wj31","country_code":"ES","type":"education","lineage":["https://openalex.org/I9617848"]}],"countries":["ES"],"is_corresponding":false,"raw_author_name":"Francesc Moll","raw_affiliation_strings":["Barcelona Supercomputing Center (BSC), Barcelona, Spain","Barcelona Supercomputing Center (BSC)Universitat Polit`ecnica de Catalunya (UPC)"],"affiliations":[{"raw_affiliation_string":"Barcelona Supercomputing Center (BSC), Barcelona, Spain","institution_ids":["https://openalex.org/I9617848","https://openalex.org/I2799803557"]},{"raw_affiliation_string":"Barcelona Supercomputing Center (BSC)Universitat Polit`ecnica de Catalunya (UPC)","institution_ids":["https://openalex.org/I2799803557","https://openalex.org/I9617848"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5020861175","display_name":"Jaume Abella","orcid":"https://orcid.org/0000-0001-7951-4028"},"institutions":[{"id":"https://openalex.org/I2799803557","display_name":"Barcelona Supercomputing Center","ror":"https://ror.org/05sd8tv96","country_code":"ES","type":"facility","lineage":["https://openalex.org/I2799803557","https://openalex.org/I9617848"]},{"id":"https://openalex.org/I9617848","display_name":"Universitat Polit\u00e8cnica de Catalunya","ror":"https://ror.org/03mb6wj31","country_code":"ES","type":"education","lineage":["https://openalex.org/I9617848"]}],"countries":["ES"],"is_corresponding":false,"raw_author_name":"Jaume Abella","raw_affiliation_strings":["Barcelona Supercomputing Center (BSC), Barcelona, Spain"],"affiliations":[{"raw_affiliation_string":"Barcelona Supercomputing Center (BSC), Barcelona, Spain","institution_ids":["https://openalex.org/I9617848","https://openalex.org/I2799803557"]}]}],"institutions":[],"countries_distinct_count":3,"institutions_distinct_count":10,"corresponding_author_ids":["https://openalex.org/A5079867677"],"corresponding_institution_ids":["https://openalex.org/I2799803557","https://openalex.org/I9617848"],"apc_list":null,"apc_paid":null,"fwci":1.0084,"has_fulltext":true,"cited_by_count":3,"citation_normalized_percentile":{"value":0.76207411,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":91,"max":99},"biblio":{"volume":"74","issue":"2","first_page":"582","last_page":"596"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11609","display_name":"Geophysical Methods and Applications","score":0.979200005531311,"subfield":{"id":"https://openalex.org/subfields/2212","display_name":"Ocean 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/T11609","display_name":"Geophysical Methods and Applications","score":0.979200005531311,"subfield":{"id":"https://openalex.org/subfields/2212","display_name":"Ocean 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/T12702","display_name":"Brain Tumor Detection and Classification","score":0.975600004196167,"subfield":{"id":"https://openalex.org/subfields/2808","display_name":"Neurology"},"field":{"id":"https://openalex.org/fields/28","display_name":"Neuroscience"},"domain":{"id":"https://openalex.org/domains/1","display_name":"Life Sciences"}},{"id":"https://openalex.org/T10201","display_name":"Speech Recognition and Synthesis","score":0.9646000266075134,"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/computer-science","display_name":"Computer science","score":0.7490083575248718},{"id":"https://openalex.org/keywords/binary-number","display_name":"Binary number","score":0.668548583984375},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.61371910572052},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.5218062400817871},{"id":"https://openalex.org/keywords/segmentation","display_name":"Segmentation","score":0.4604772627353668},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.3982222378253937},{"id":"https://openalex.org/keywords/arithmetic","display_name":"Arithmetic","score":0.2682579755783081},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.2538958787918091},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.16045284271240234}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7490083575248718},{"id":"https://openalex.org/C48372109","wikidata":"https://www.wikidata.org/wiki/Q3913","display_name":"Binary number","level":2,"score":0.668548583984375},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.61371910572052},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.5218062400817871},{"id":"https://openalex.org/C89600930","wikidata":"https://www.wikidata.org/wiki/Q1423946","display_name":"Segmentation","level":2,"score":0.4604772627353668},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.3982222378253937},{"id":"https://openalex.org/C94375191","wikidata":"https://www.wikidata.org/wiki/Q11205","display_name":"Arithmetic","level":1,"score":0.2682579755783081},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.2538958787918091},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.16045284271240234}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1109/tc.2024.3500369","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tc.2024.3500369","pdf_url":null,"source":{"id":"https://openalex.org/S157670870","display_name":"IEEE Transactions on Computers","issn_l":"0018-9340","issn":["0018-9340","1557-9956","2326-3814"],"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 Computers","raw_type":"journal-article"},{"id":"pmh:oai:upcommons.upc.edu:2117/426140","is_oa":true,"landing_page_url":"https://hdl.handle.net/2117/426140","pdf_url":"https://upcommons.upc.edu/bitstreams/1cca3fd2-04ca-4d45-b44b-b7fdf16f2358/download","source":{"id":"https://openalex.org/S4377196262","display_name":"UPCommons institutional repository (Universitat Polit\u00e8cnica de Catalunya)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I9617848","host_organization_name":"Universitat Polit\u00e8cnica de Catalunya","host_organization_lineage":["https://openalex.org/I9617848"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by-nc-nd","license_id":"https://openalex.org/licenses/cc-by-nc-nd","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"info:eu-repo/semantics/article"}],"best_oa_location":{"id":"pmh:oai:upcommons.upc.edu:2117/426140","is_oa":true,"landing_page_url":"https://hdl.handle.net/2117/426140","pdf_url":"https://upcommons.upc.edu/bitstreams/1cca3fd2-04ca-4d45-b44b-b7fdf16f2358/download","source":{"id":"https://openalex.org/S4377196262","display_name":"UPCommons institutional repository (Universitat Polit\u00e8cnica de Catalunya)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I9617848","host_organization_name":"Universitat Polit\u00e8cnica de Catalunya","host_organization_lineage":["https://openalex.org/I9617848"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by-nc-nd","license_id":"https://openalex.org/licenses/cc-by-nc-nd","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"info:eu-repo/semantics/article"},"sustainable_development_goals":[{"display_name":"Affordable and clean energy","score":0.8500000238418579,"id":"https://metadata.un.org/sdg/7"}],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4404609150.pdf","grobid_xml":"https://content.openalex.org/works/W4404609150.grobid-xml"},"referenced_works_count":31,"referenced_works":["https://openalex.org/W152964218","https://openalex.org/W2011136852","https://openalex.org/W2194775991","https://openalex.org/W2252007067","https://openalex.org/W2618530766","https://openalex.org/W2889339831","https://openalex.org/W2945146780","https://openalex.org/W2963125010","https://openalex.org/W2963255460","https://openalex.org/W2970796375","https://openalex.org/W3012561096","https://openalex.org/W3013186616","https://openalex.org/W3032746325","https://openalex.org/W3036554530","https://openalex.org/W3047342117","https://openalex.org/W3049586678","https://openalex.org/W3118434062","https://openalex.org/W3137147200","https://openalex.org/W3139307480","https://openalex.org/W3156145546","https://openalex.org/W3195130375","https://openalex.org/W3202361117","https://openalex.org/W3204960814","https://openalex.org/W4206957512","https://openalex.org/W4312443924","https://openalex.org/W4313484738","https://openalex.org/W4360831846","https://openalex.org/W4387319216","https://openalex.org/W6637373629","https://openalex.org/W6737664043","https://openalex.org/W6762718338"],"related_works":["https://openalex.org/W2051487156","https://openalex.org/W2073681303","https://openalex.org/W2053286651","https://openalex.org/W2055243143","https://openalex.org/W2181743346","https://openalex.org/W2187401768","https://openalex.org/W2181413294","https://openalex.org/W2989452537","https://openalex.org/W2052122378","https://openalex.org/W2544423928"],"abstract_inverted_index":{"Efficiently":[0],"computing":[1],"Deep":[2],"Neural":[3],"Networks":[4],"(DNNs)":[5],"has":[6,32],"become":[7],"a":[8,89,133,149,158],"primary":[9],"challenge":[10],"in":[11,42,132,157],"today's":[12],"computers,":[13],"especially":[14],"on":[15,24,72],"devices":[16],"targeting":[17],"mobile":[18],"or":[19,52],"edge":[20],"applications.":[21],"Recent":[22],"progress":[23],"Post-Training":[25],"Quantization":[26],"(PTQ)":[27],"and":[28,64,141,172,180,209],"Quantization-Aware":[29],"Training":[30],"(QAT)":[31],"shown":[33],"that":[34,93,164],"the":[35,73,125,129,186,213,220],"key":[36],"to":[37,50,55,79,97,111,143,168,177,185,201],"high":[38,145],"energy":[39,146,173,210],"efficiency":[40,147,174,211],"lies":[41],"executing":[43],"deep":[44],"learning":[45],"models":[46],"with":[47,183,189,205],"low-":[48],"(8-":[49],"5-bit)":[51],"ultra-low-precision":[53],"(4-":[54],"2-bit).":[56],"Unfortunately,":[57],"current":[58],"Central":[59],"Processing":[60],"Unit":[61],"(CPU)":[62],"architectures":[63],"Instruction":[65],"Set":[66],"Architectures":[67],"(ISAs)":[68],"present":[69,87],"severe":[70],"limitations":[71],"range":[74],"of":[75,128,137,193],"data":[76,106,126],"sizes":[77],"supported":[78],"compute":[80,99],"DNN":[81,102],"kernels.":[82],"In":[83],"this":[84],"work,":[85],"we":[86,165],"<i>Mix-GEMM</i>,":[88],"hardware-software":[90],"co-designed":[91],"architecture":[92,118,156],"enables":[94],"RISC-V":[95,161],"processors":[96],"efficiently":[98],"arbitrary":[100],"mixed-precision":[101],"kernels,":[103],"supporting":[104],"all":[105],"size":[107,127],"combinations":[108],"from":[109],"8-":[110],"2-bit.":[112],"By":[113],"applying":[114],"<i>binary":[115],"segmentation</i>,":[116],"our":[117,154,198],"can":[119],"scale":[120],"its":[121,170],"throughput":[122],"by":[123,175],"decreasing":[124],"operands,":[130],"resulting":[131],"flexible":[134],"approach":[135],"capable":[136],"leveraging":[138],"state-of-the-art":[139,203],"QAT":[140],"PTQ":[142],"achieve":[144],"at":[148],"very":[150],"low":[151],"cost.":[152],"Evaluating":[153],"<i>Mix-GEMM</i>":[155],"dual-issue":[159],"in-order":[160],"processor":[162,200],"shows":[163],"are":[166],"able":[167],"boost":[169],"performance":[171,208],"up":[176],"<inline-formula><tex-math":[178,181],"notation=\"LaTeX\">$44\\times$</tex-math></inline-formula>":[179],"notation=\"LaTeX\">$11\\times$</tex-math></inline-formula>":[182],"respect":[184],"baseline":[187],"processor,":[188],"an":[190],"area":[191],"overhead":[192],"only":[194],"2%.":[195],"This":[196],"allows":[197],"extended":[199],"execute":[202],"DNNs":[204],"significantly":[206],"higher":[207],"than":[212],"standard":[214],"FP32":[215],"precision,":[216],"while":[217],"retaining":[218],"almost":[219],"same":[221],"model":[222],"accuracy.":[223]},"counts_by_year":[{"year":2026,"cited_by_count":2},{"year":2025,"cited_by_count":1}],"updated_date":"2026-03-27T05:58:40.876381","created_date":"2025-10-10T00:00:00"}