{"id":"https://openalex.org/W7167740293","doi":"https://doi.org/10.1145/3811257.3811258","title":"Optimizing Hardware Utilization and Simplifying Development for AI Kernels","display_name":"Optimizing Hardware Utilization and Simplifying Development for AI Kernels","publication_year":2026,"publication_date":"2026-05-06","ids":{"openalex":"https://openalex.org/W7167740293","doi":"https://doi.org/10.1145/3811257.3811258"},"language":null,"primary_location":{"id":"doi:10.1145/3811257.3811258","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3811257.3811258","pdf_url":null,"source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the International Workshop on OpenCL and SYCL","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://doi.org/10.1145/3811257.3811258","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5140262721","display_name":"Feng Yuan","orcid":null},"institutions":[{"id":"https://openalex.org/I1343180700","display_name":"Intel (United States)","ror":"https://ror.org/01ek73717","country_code":"US","type":"company","lineage":["https://openalex.org/I1343180700"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Feng Yuan","raw_affiliation_strings":["Intel China, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Intel China, China","institution_ids":["https://openalex.org/I1343180700"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5039391307","display_name":"Jinghui Gu","orcid":null},"institutions":[{"id":"https://openalex.org/I1343180700","display_name":"Intel (United States)","ror":"https://ror.org/01ek73717","country_code":"US","type":"company","lineage":["https://openalex.org/I1343180700"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Jinghui Gu","raw_affiliation_strings":["Intel China, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Intel China, China","institution_ids":["https://openalex.org/I1343180700"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5010063789","display_name":"Yejun Guo","orcid":null},"institutions":[{"id":"https://openalex.org/I1343180700","display_name":"Intel (United States)","ror":"https://ror.org/01ek73717","country_code":"US","type":"company","lineage":["https://openalex.org/I1343180700"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Yejun Guo","raw_affiliation_strings":["Intel China, Shanghai, China"],"raw_orcid":"https://orcid.org/0009-0000-1072-9477","affiliations":[{"raw_affiliation_string":"Intel China, Shanghai, China","institution_ids":["https://openalex.org/I1343180700"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5102000779","display_name":"Jian Qiu Zhang","orcid":"https://orcid.org/0000-0003-3369-1461"},"institutions":[{"id":"https://openalex.org/I1343180700","display_name":"Intel (United States)","ror":"https://ror.org/01ek73717","country_code":"US","type":"company","lineage":["https://openalex.org/I1343180700"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Jian Zhang","raw_affiliation_strings":["Intel China, Shanghai, China"],"raw_orcid":"https://orcid.org/0000-0003-3369-1461","affiliations":[{"raw_affiliation_string":"Intel China, Shanghai, China","institution_ids":["https://openalex.org/I1343180700"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I1343180700"],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"2"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.8406000137329102,"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":0.8406000137329102,"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/T10036","display_name":"Advanced Neural Network Applications","score":0.0406000018119812,"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/T10904","display_name":"Embedded Systems Design Techniques","score":0.02410000003874302,"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/kernel","display_name":"Kernel (algebra)","score":0.6122000217437744},{"id":"https://openalex.org/keywords/workgroup","display_name":"Workgroup","score":0.5295000076293945},{"id":"https://openalex.org/keywords/concurrency","display_name":"Concurrency","score":0.4733999967575073},{"id":"https://openalex.org/keywords/scheme","display_name":"Scheme (mathematics)","score":0.38690000772476196},{"id":"https://openalex.org/keywords/implementation","display_name":"Implementation","score":0.36239999532699585},{"id":"https://openalex.org/keywords/matching","display_name":"Matching (statistics)","score":0.3425999879837036},{"id":"https://openalex.org/keywords/thread","display_name":"Thread (computing)","score":0.31189998984336853},{"id":"https://openalex.org/keywords/host","display_name":"Host (biology)","score":0.3100000023841858}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8348000049591064},{"id":"https://openalex.org/C74193536","wikidata":"https://www.wikidata.org/wiki/Q574844","display_name":"Kernel (algebra)","level":2,"score":0.6122000217437744},{"id":"https://openalex.org/C95423123","wikidata":"https://www.wikidata.org/wiki/Q622178","display_name":"Workgroup","level":2,"score":0.5295000076293945},{"id":"https://openalex.org/C118524514","wikidata":"https://www.wikidata.org/wiki/Q173212","display_name":"Computer architecture","level":1,"score":0.4934000074863434},{"id":"https://openalex.org/C193702766","wikidata":"https://www.wikidata.org/wiki/Q1414548","display_name":"Concurrency","level":2,"score":0.4733999967575073},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.4350999891757965},{"id":"https://openalex.org/C77618280","wikidata":"https://www.wikidata.org/wiki/Q1155772","display_name":"Scheme (mathematics)","level":2,"score":0.38690000772476196},{"id":"https://openalex.org/C26713055","wikidata":"https://www.wikidata.org/wiki/Q245962","display_name":"Implementation","level":2,"score":0.36239999532699585},{"id":"https://openalex.org/C165064840","wikidata":"https://www.wikidata.org/wiki/Q1321061","display_name":"Matching (statistics)","level":2,"score":0.3425999879837036},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.3206999897956848},{"id":"https://openalex.org/C149635348","wikidata":"https://www.wikidata.org/wiki/Q193040","display_name":"Embedded system","level":1,"score":0.31290000677108765},{"id":"https://openalex.org/C138101251","wikidata":"https://www.wikidata.org/wiki/Q213092","display_name":"Thread (computing)","level":2,"score":0.31189998984336853},{"id":"https://openalex.org/C126831891","wikidata":"https://www.wikidata.org/wiki/Q221673","display_name":"Host (biology)","level":2,"score":0.3100000023841858},{"id":"https://openalex.org/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.29589998722076416},{"id":"https://openalex.org/C75165309","wikidata":"https://www.wikidata.org/wiki/Q2258979","display_name":"Search engine indexing","level":2,"score":0.2874000072479248},{"id":"https://openalex.org/C142962650","wikidata":"https://www.wikidata.org/wiki/Q240838","display_name":"Reconfigurable computing","level":3,"score":0.28700000047683716},{"id":"https://openalex.org/C22174128","wikidata":"https://www.wikidata.org/wiki/Q175869","display_name":"Microcode","level":2,"score":0.28349998593330383},{"id":"https://openalex.org/C34165917","wikidata":"https://www.wikidata.org/wiki/Q188267","display_name":"Programming paradigm","level":2,"score":0.2800999879837036},{"id":"https://openalex.org/C65232700","wikidata":"https://www.wikidata.org/wiki/Q5656403","display_name":"Hardware architecture","level":3,"score":0.2782000005245209},{"id":"https://openalex.org/C2777904410","wikidata":"https://www.wikidata.org/wiki/Q7397","display_name":"Software","level":2,"score":0.27799999713897705},{"id":"https://openalex.org/C63000827","wikidata":"https://www.wikidata.org/wiki/Q3080428","display_name":"Software portability","level":2,"score":0.2768000066280365},{"id":"https://openalex.org/C192209626","wikidata":"https://www.wikidata.org/wiki/Q190909","display_name":"Focus (optics)","level":2,"score":0.2743000090122223},{"id":"https://openalex.org/C177225278","wikidata":"https://www.wikidata.org/wiki/Q192674","display_name":"Factoring","level":2,"score":0.27149999141693115},{"id":"https://openalex.org/C58013763","wikidata":"https://www.wikidata.org/wiki/Q5754574","display_name":"High-level synthesis","level":3,"score":0.27140000462532043},{"id":"https://openalex.org/C157170001","wikidata":"https://www.wikidata.org/wiki/Q4781507","display_name":"Applications of artificial intelligence","level":2,"score":0.2639000117778778},{"id":"https://openalex.org/C9390403","wikidata":"https://www.wikidata.org/wiki/Q3966","display_name":"Computer hardware","level":1,"score":0.2547000050544739},{"id":"https://openalex.org/C96147967","wikidata":"https://www.wikidata.org/wiki/Q190686","display_name":"Subroutine","level":2,"score":0.25119999051094055}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3811257.3811258","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3811257.3811258","pdf_url":null,"source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the International Workshop on OpenCL and SYCL","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.1145/3811257.3811258","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3811257.3811258","pdf_url":null,"source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the International Workshop on OpenCL and SYCL","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Traditional":[0],"SIMT-based":[1],"parallel":[2],"programming":[3],"languages":[4,35],"(including":[5],"SYCL,":[6],"CUDA,":[7],"and":[8,21,53,91,99,103,145,158],"OpenCL)":[9],"are":[10],"widely":[11],"adopted":[12],"for":[13,149],"implementing":[14],"high-performance":[15],"AI":[16,118],"kernels,":[17],"as":[18,153],"their":[19],"thread":[20],"workgroup":[22,93],"abstractions":[23],"enable":[24],"fine-grained":[25],"hardware":[26,51,68,89,129],"resource":[27],"utilization.":[28],"However,":[29],"developing":[30],"efficient":[31],"kernels":[32],"with":[33,167],"these":[34],"imposes":[36],"heavy":[37],"burdens":[38],"on":[39,95,124],"developers:":[40],"they":[41],"must":[42],"master":[43],"kernel":[44,72,119,165],"algorithm":[45,73,125],"design,":[46],"deeply":[47],"understand":[48],"target":[49],"GPU":[50,140],"architectures,":[52],"implement":[54],"low-level":[55],"hardware-aware":[56],"optimizations":[57],"manually.":[58],"This":[59],"paper":[60],"presents":[61],"an":[62],"adaptive":[63,92],"optimization":[64],"framework":[65,86,112],"that":[66],"decouples":[67],"utilization":[69],"tuning":[70],"from":[71,79],"development,":[74],"hiding":[75],"complex":[76],"hardware-specific":[77],"details":[78],"developers":[80,120],"while":[81],"delivering":[82],"peak":[83],"performance.":[84],"The":[85,135],"performs":[87],"automatic":[88],"detection":[90],"dispatching":[94],"the":[96,107,114],"host":[97],"side,":[98],"introduces":[100],"kernel-semantics-oriented":[101],"memory":[102],"concurrency":[104],"mapping":[105],"inside":[106],"kernel.":[108],"By":[109],"integrating":[110],"this":[111],"into":[113],"PyTorch":[115],"XPU":[116],"backend,":[117],"can":[121],"focus":[122],"exclusively":[123],"logic":[126],"without":[127],"learning":[128],"architectures":[130,141],"or":[131],"hardware-software":[132],"co-design":[133],"techniques.":[134],"optimized":[136],"scheme":[137],"supports":[138],"mainstream":[139],"(Intel,":[142],"NVIDIA,":[143],"AMD)":[144],"has":[146],"been":[147],"deployed":[148],"common":[150],"operators":[151],"such":[152],"MaxPooling,":[154],"WeightNormalization,":[155],"Scan,":[156],"EmbeddingBag,":[157],"indexing":[159],"operations,":[160],"achieving":[161],"performance":[162],"matching":[163],"hand-tuned":[164],"implementations":[166],"drastically":[168],"simplified":[169],"development":[170],"workflows.":[171]},"counts_by_year":[],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2026-07-09T00:00:00"}