{"id":"https://openalex.org/W7128300996","doi":"https://doi.org/10.26599/bdma.2025.9020065","title":"Leveraging Large-Scale Data for Efficient Low-Bit CUTLASS GEMM Optimization via Neural Networks","display_name":"Leveraging Large-Scale Data for Efficient Low-Bit CUTLASS GEMM Optimization via Neural Networks","publication_year":2026,"publication_date":"2026-02-06","ids":{"openalex":"https://openalex.org/W7128300996","doi":"https://doi.org/10.26599/bdma.2025.9020065"},"language":"en","primary_location":{"id":"doi:10.26599/bdma.2025.9020065","is_oa":true,"landing_page_url":"https://doi.org/10.26599/bdma.2025.9020065","pdf_url":null,"source":{"id":"https://openalex.org/S4210209060","display_name":"Big Data Mining and Analytics","issn_l":"2096-0654","issn":["2096-0654","2097-406X"],"is_oa":true,"is_in_doaj":true,"is_core":true,"host_organization":"https://openalex.org/P4310311901","host_organization_name":"Tsinghua University Press","host_organization_lineage":["https://openalex.org/P4310311901"],"host_organization_lineage_names":["Tsinghua University Press"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Big Data Mining and Analytics","raw_type":"journal-article"},"type":"article","indexed_in":["crossref","doaj"],"open_access":{"is_oa":true,"oa_status":"diamond","oa_url":"https://doi.org/10.26599/bdma.2025.9020065","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5125358975","display_name":"Hong Guo","orcid":null},"institutions":[{"id":"https://openalex.org/I143288331","display_name":"Hasso Plattner Institute","ror":"https://ror.org/058rn5r42","country_code":"DE","type":"facility","lineage":["https://openalex.org/I143288331","https://openalex.org/I176453806"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Hong Guo","raw_affiliation_strings":["Hasso Plattner Institute for Digital Engineering gGmbH, University of Potsdam,Potsdam,Germany,14482"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Hasso Plattner Institute for Digital Engineering gGmbH, University of Potsdam,Potsdam,Germany,14482","institution_ids":["https://openalex.org/I143288331"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5043579361","display_name":"Nianhui Guo","orcid":null},"institutions":[{"id":"https://openalex.org/I143288331","display_name":"Hasso Plattner Institute","ror":"https://ror.org/058rn5r42","country_code":"DE","type":"facility","lineage":["https://openalex.org/I143288331","https://openalex.org/I176453806"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Nianhui Guo","raw_affiliation_strings":["Hasso Plattner Institute for Digital Engineering gGmbH, University of Potsdam,Potsdam,Germany,14482"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Hasso Plattner Institute for Digital Engineering gGmbH, University of Potsdam,Potsdam,Germany,14482","institution_ids":["https://openalex.org/I143288331"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5033952850","display_name":"Christoph Meinel","orcid":"https://orcid.org/0000-0002-3212-6201"},"institutions":[{"id":"https://openalex.org/I143288331","display_name":"Hasso Plattner Institute","ror":"https://ror.org/058rn5r42","country_code":"DE","type":"facility","lineage":["https://openalex.org/I143288331","https://openalex.org/I176453806"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Christoph Meinel","raw_affiliation_strings":["Hasso Plattner Institute for Digital Engineering gGmbH, University of Potsdam,Potsdam,Germany,14482"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Hasso Plattner Institute for Digital Engineering gGmbH, University of Potsdam,Potsdam,Germany,14482","institution_ids":["https://openalex.org/I143288331"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5101039243","display_name":"Haojin Yang","orcid":null},"institutions":[{"id":"https://openalex.org/I143288331","display_name":"Hasso Plattner Institute","ror":"https://ror.org/058rn5r42","country_code":"DE","type":"facility","lineage":["https://openalex.org/I143288331","https://openalex.org/I176453806"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Haojin Yang","raw_affiliation_strings":["Hasso Plattner Institute for Digital Engineering gGmbH, University of Potsdam,Potsdam,Germany,14482"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Hasso Plattner Institute for Digital Engineering gGmbH, University of Potsdam,Potsdam,Germany,14482","institution_ids":["https://openalex.org/I143288331"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I143288331"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.09506871,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":"9","issue":"2","first_page":"632","last_page":"652"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.13600000739097595,"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/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.13600000739097595,"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/T11612","display_name":"Stochastic Gradient Optimization Techniques","score":0.05209999904036522,"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/T11697","display_name":"Numerical Methods and Algorithms","score":0.03790000081062317,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.6026999950408936},{"id":"https://openalex.org/keywords/stability","display_name":"Stability (learning theory)","score":0.33309999108314514},{"id":"https://openalex.org/keywords/feature","display_name":"Feature (linguistics)","score":0.30079999566078186},{"id":"https://openalex.org/keywords/deep-neural-networks","display_name":"Deep neural networks","score":0.27549999952316284},{"id":"https://openalex.org/keywords/key","display_name":"Key (lock)","score":0.26930001378059387}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6984000205993652},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.6026999950408936},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.49889999628067017},{"id":"https://openalex.org/C112972136","wikidata":"https://www.wikidata.org/wiki/Q7595718","display_name":"Stability (learning theory)","level":2,"score":0.33309999108314514},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.3319999873638153},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3312999904155731},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.30640000104904175},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.30079999566078186},{"id":"https://openalex.org/C2984842247","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep neural networks","level":3,"score":0.27549999952316284},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.26930001378059387},{"id":"https://openalex.org/C147764199","wikidata":"https://www.wikidata.org/wiki/Q6865248","display_name":"Minification","level":2,"score":0.2572999894618988}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.26599/bdma.2025.9020065","is_oa":true,"landing_page_url":"https://doi.org/10.26599/bdma.2025.9020065","pdf_url":null,"source":{"id":"https://openalex.org/S4210209060","display_name":"Big Data Mining and Analytics","issn_l":"2096-0654","issn":["2096-0654","2097-406X"],"is_oa":true,"is_in_doaj":true,"is_core":true,"host_organization":"https://openalex.org/P4310311901","host_organization_name":"Tsinghua University Press","host_organization_lineage":["https://openalex.org/P4310311901"],"host_organization_lineage_names":["Tsinghua University Press"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Big Data Mining and Analytics","raw_type":"journal-article"},{"id":"pmh:oai:doaj.org/article:ab4f1c7956d146baa8c52f874a52bef1","is_oa":true,"landing_page_url":"https://doaj.org/article/ab4f1c7956d146baa8c52f874a52bef1","pdf_url":null,"source":{"id":"https://openalex.org/S4306401280","display_name":"DOAJ (DOAJ: Directory of Open Access Journals)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by-sa","license_id":"https://openalex.org/licenses/cc-by-sa","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"Big Data Mining and Analytics, Vol 9, Iss 2, Pp 632-652 (2026)","raw_type":"article"}],"best_oa_location":{"id":"doi:10.26599/bdma.2025.9020065","is_oa":true,"landing_page_url":"https://doi.org/10.26599/bdma.2025.9020065","pdf_url":null,"source":{"id":"https://openalex.org/S4210209060","display_name":"Big Data Mining and Analytics","issn_l":"2096-0654","issn":["2096-0654","2097-406X"],"is_oa":true,"is_in_doaj":true,"is_core":true,"host_organization":"https://openalex.org/P4310311901","host_organization_name":"Tsinghua University Press","host_organization_lineage":["https://openalex.org/P4310311901"],"host_organization_lineage_names":["Tsinghua University Press"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Big Data Mining and Analytics","raw_type":"journal-article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":25,"referenced_works":["https://openalex.org/W1863336885","https://openalex.org/W1978642402","https://openalex.org/W1988115241","https://openalex.org/W2100115659","https://openalex.org/W2103496339","https://openalex.org/W2132083787","https://openalex.org/W2804032941","https://openalex.org/W2885184702","https://openalex.org/W2901312569","https://openalex.org/W2903669223","https://openalex.org/W2913790721","https://openalex.org/W2984035546","https://openalex.org/W3014127403","https://openalex.org/W3117765988","https://openalex.org/W3120016685","https://openalex.org/W3177265267","https://openalex.org/W4312933868","https://openalex.org/W4316252370","https://openalex.org/W4390097705","https://openalex.org/W4393307967","https://openalex.org/W4400914074","https://openalex.org/W4403769934","https://openalex.org/W4403794376","https://openalex.org/W4408151434","https://openalex.org/W4409363239"],"related_works":[],"abstract_inverted_index":{"Optimizing":[0],"GEneral":[1],"Matrix":[2],"Multiplication":[3],"(GEMM)":[4],"on":[5,61,101,123,140,151],"GPU":[6,68,154],"platforms":[7],"is":[8,99,120],"becoming":[9],"increasingly":[10],"critical":[11],"to":[12,88,107,138,159,171],"meet":[13],"the":[14,34,141,152,167,172],"growing":[15],"computational":[16],"demands":[17],"of":[18,36,136,147,157],"modern":[19],"deep":[20],"neural":[21,85],"network":[22,86],"research.":[23],"While":[24],"significant":[25],"progress":[26],"has":[27],"been":[28],"made":[29],"in":[30],"accelerating":[31],"high-precision":[32],"GEMM,":[33,82],"optimization":[35],"low-bit":[37,50,80,148],"GEMM":[38,51,91,162],"remains":[39],"a":[40,75,84,102],"challenging":[41],"open":[42],"problem.":[43],"The":[44],"CUTLASS":[45,81,174],"library":[46],"provides":[47],"highly":[48],"optimized":[49],"templates":[52],"leveraging":[53],"Tensor":[54],"Cores;":[55],"however,":[56],"performance":[57],"varies":[58],"considerably":[59],"depending":[60],"tile":[62],"and":[63,119,164],"pipeline":[64],"configurations":[65],"across":[66,115],"different":[67],"architectures.":[69],"In":[70],"this":[71],"work,":[72],"we":[73],"propose":[74],"novel":[76],"auto-tuning":[77],"framework":[78],"for":[79,94,161,166],"utilizing":[83],"model":[87,98],"predict":[89],"optimal":[90],"template":[92],"parameters":[93],"target":[95],"GPUs.":[96],"Our":[97],"trained":[100],"synthetic":[103],"dataset":[104],"with":[105],"up":[106,137,158],"116100":[108],"unique":[109],"samples,":[110],"encompassing":[111],"diverse":[112],"matrix":[113],"sizes":[114],"various":[116],"Ampere":[117],"GPUs,":[118],"thoroughly":[121],"evaluated":[122],"these":[124],"hardware":[125],"platforms.":[126],"Experimental":[127],"results":[128],"show":[129],"that":[130],"our":[131],"method":[132],"achieves":[133],"an":[134],"accuracy":[135],"95.11%":[139],"validation":[142],"dataset.":[143],"Furthermore,":[144],"real-time":[145],"evaluations":[146],"data":[149],"types":[150],"A100":[153],"demonstrate":[155],"speedups":[156],"1.99\u00d7":[160],"operations":[163],"1.28\u00d7":[165],"linear":[168],"layer,":[169],"compared":[170],"default":[173],"templates.":[175]},"counts_by_year":[],"updated_date":"2026-02-09T06:03:36.439944","created_date":"2026-02-09T00:00:00"}