{"id":"https://openalex.org/W4417241966","doi":"https://doi.org/10.1145/3760250.3762219","title":"Tilus: A Tile-Level GPGPU Programming Language for Low-Precision Computation","display_name":"Tilus: A Tile-Level GPGPU Programming Language for Low-Precision Computation","publication_year":2025,"publication_date":"2025-12-11","ids":{"openalex":"https://openalex.org/W4417241966","doi":"https://doi.org/10.1145/3760250.3762219"},"language":null,"primary_location":{"id":"doi:10.1145/3760250.3762219","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3760250.3762219","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3760250.3762219","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 31st ACM International Conference on Architectural Support for Programming Languages and Operating Systems, Volume 1","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://dl.acm.org/doi/pdf/10.1145/3760250.3762219","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5015556239","display_name":"Yaoyao Ding","orcid":"https://orcid.org/0000-0002-9801-6080"},"institutions":[{"id":"https://openalex.org/I4210127509","display_name":"Vector Institute","ror":"https://ror.org/03kqdja62","country_code":"CA","type":"facility","lineage":["https://openalex.org/I4210127509"]}],"countries":["CA"],"is_corresponding":true,"raw_author_name":"Yaoyao Ding","raw_affiliation_strings":["Univerisity of Toronto, Toronto, ON, Canada, NVIDIA, Santa Clara, CA, USA, and Vector Institute, Toronto, ON, Canada"],"affiliations":[{"raw_affiliation_string":"Univerisity of Toronto, Toronto, ON, Canada, NVIDIA, Santa Clara, CA, USA, and Vector Institute, Toronto, ON, Canada","institution_ids":["https://openalex.org/I4210127509"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5065994853","display_name":"Bohan Hou","orcid":"https://orcid.org/0000-0001-5718-3387"},"institutions":[{"id":"https://openalex.org/I74973139","display_name":"Carnegie Mellon University","ror":"https://ror.org/05x2bcf33","country_code":"US","type":"education","lineage":["https://openalex.org/I74973139"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Bohan Hou","raw_affiliation_strings":["Carnegie Mellon University, Pittsburgh, PA, USA"],"affiliations":[{"raw_affiliation_string":"Carnegie Mellon University, Pittsburgh, PA, USA","institution_ids":["https://openalex.org/I74973139"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5108698354","display_name":"Xiao Zhang","orcid":"https://orcid.org/0009-0000-0914-5669"},"institutions":[{"id":"https://openalex.org/I1304085615","display_name":"Nvidia (United Kingdom)","ror":"https://ror.org/02kr42612","country_code":"GB","type":"company","lineage":["https://openalex.org/I1304085615","https://openalex.org/I4210127875"]},{"id":"https://openalex.org/I4210127875","display_name":"Nvidia (United States)","ror":"https://ror.org/03jdj4y14","country_code":"US","type":"company","lineage":["https://openalex.org/I4210127875"]}],"countries":["GB","US"],"is_corresponding":false,"raw_author_name":"Xiao Zhang","raw_affiliation_strings":["University of Toronto, Toronto, ON, Canada and NVIDIA, Santa Clara, CA, Canada"],"affiliations":[{"raw_affiliation_string":"University of Toronto, Toronto, ON, Canada and NVIDIA, Santa Clara, CA, Canada","institution_ids":["https://openalex.org/I4210127875","https://openalex.org/I1304085615"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5111210524","display_name":"A. Binus Y. Lin","orcid":"https://orcid.org/0009-0008-0401-1053"},"institutions":[{"id":"https://openalex.org/I151746483","display_name":"University of Waterloo","ror":"https://ror.org/01aff2v68","country_code":"CA","type":"education","lineage":["https://openalex.org/I151746483"]}],"countries":["CA"],"is_corresponding":false,"raw_author_name":"Allan Lin","raw_affiliation_strings":["University of Waterloo, Waterloo, ON, Canada"],"affiliations":[{"raw_affiliation_string":"University of Waterloo, Waterloo, ON, Canada","institution_ids":["https://openalex.org/I151746483"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5023529355","display_name":"Tianqi Chen","orcid":"https://orcid.org/0000-0002-5744-3940"},"institutions":[{"id":"https://openalex.org/I74973139","display_name":"Carnegie Mellon University","ror":"https://ror.org/05x2bcf33","country_code":"US","type":"education","lineage":["https://openalex.org/I74973139"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Tianqi Chen","raw_affiliation_strings":["Carnegie Mellon University, Pittsburgh, PA, USA and NVIDIA, Santa Clara, CA, USA"],"affiliations":[{"raw_affiliation_string":"Carnegie Mellon University, Pittsburgh, PA, USA and NVIDIA, Santa Clara, CA, USA","institution_ids":["https://openalex.org/I74973139"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5008633310","display_name":"Cody Hao Yu","orcid":"https://orcid.org/0000-0002-9298-6254"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Cody Hao Yu","raw_affiliation_strings":["Independent Researcher, Santa Clara, CA, USA"],"affiliations":[{"raw_affiliation_string":"Independent Researcher, Santa Clara, CA, USA","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101638214","display_name":"Yida Wang","orcid":"https://orcid.org/0000-0001-8165-840X"},"institutions":[{"id":"https://openalex.org/I1311688040","display_name":"Amazon (United States)","ror":"https://ror.org/04mv4n011","country_code":"US","type":"company","lineage":["https://openalex.org/I1311688040"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Yida Wang","raw_affiliation_strings":["Amazon Web Services, Santa Clara, CA, USA"],"affiliations":[{"raw_affiliation_string":"Amazon Web Services, Santa Clara, CA, USA","institution_ids":["https://openalex.org/I1311688040"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5007585346","display_name":"Gennady Pekhimenko","orcid":"https://orcid.org/0000-0002-3839-0919"},"institutions":[{"id":"https://openalex.org/I4210127509","display_name":"Vector Institute","ror":"https://ror.org/03kqdja62","country_code":"CA","type":"facility","lineage":["https://openalex.org/I4210127509"]}],"countries":["CA"],"is_corresponding":false,"raw_author_name":"Gennady Pekhimenko","raw_affiliation_strings":["University of Toronto, Toronto, ON, Canada, NVIDIA, Santa Clara, CA, Canada, and Vector Institute, Toronto, ON, Canada"],"affiliations":[{"raw_affiliation_string":"University of Toronto, Toronto, ON, Canada, NVIDIA, Santa Clara, CA, Canada, and Vector Institute, Toronto, ON, Canada","institution_ids":["https://openalex.org/I4210127509"]}]}],"institutions":[],"countries_distinct_count":3,"institutions_distinct_count":8,"corresponding_author_ids":["https://openalex.org/A5015556239"],"corresponding_institution_ids":["https://openalex.org/I4210127509"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":true,"cited_by_count":0,"citation_normalized_percentile":{"value":0.43520451,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"281","last_page":"297"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.8920999765396118,"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":0.8920999765396118,"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/T10181","display_name":"Natural Language Processing Techniques","score":0.01080000028014183,"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/T10904","display_name":"Embedded Systems Design Techniques","score":0.009100000374019146,"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/compiler","display_name":"Compiler","score":0.659500002861023},{"id":"https://openalex.org/keywords/general-purpose-computing-on-graphics-processing-units","display_name":"General-purpose computing on graphics processing units","score":0.5281000137329102},{"id":"https://openalex.org/keywords/computation","display_name":"Computation","score":0.4936999976634979},{"id":"https://openalex.org/keywords/memory-bandwidth","display_name":"Memory bandwidth","score":0.4668999910354614},{"id":"https://openalex.org/keywords/programming-paradigm","display_name":"Programming paradigm","score":0.45570001006126404},{"id":"https://openalex.org/keywords/cuda","display_name":"CUDA","score":0.4489000141620636},{"id":"https://openalex.org/keywords/key","display_name":"Key (lock)","score":0.3815000057220459},{"id":"https://openalex.org/keywords/memory-hierarchy","display_name":"Memory hierarchy","score":0.3749000132083893},{"id":"https://openalex.org/keywords/high-level-programming-language","display_name":"High-level programming language","score":0.3580999970436096}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.9021999835968018},{"id":"https://openalex.org/C169590947","wikidata":"https://www.wikidata.org/wiki/Q47506","display_name":"Compiler","level":2,"score":0.659500002861023},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.6380000114440918},{"id":"https://openalex.org/C50630238","wikidata":"https://www.wikidata.org/wiki/Q971505","display_name":"General-purpose computing on graphics processing units","level":3,"score":0.5281000137329102},{"id":"https://openalex.org/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.49399998784065247},{"id":"https://openalex.org/C45374587","wikidata":"https://www.wikidata.org/wiki/Q12525525","display_name":"Computation","level":2,"score":0.4936999976634979},{"id":"https://openalex.org/C188045654","wikidata":"https://www.wikidata.org/wiki/Q17148339","display_name":"Memory bandwidth","level":2,"score":0.4668999910354614},{"id":"https://openalex.org/C34165917","wikidata":"https://www.wikidata.org/wiki/Q188267","display_name":"Programming paradigm","level":2,"score":0.45570001006126404},{"id":"https://openalex.org/C2778119891","wikidata":"https://www.wikidata.org/wiki/Q477690","display_name":"CUDA","level":2,"score":0.4489000141620636},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.3815000057220459},{"id":"https://openalex.org/C2778100165","wikidata":"https://www.wikidata.org/wiki/Q1589327","display_name":"Memory hierarchy","level":3,"score":0.3749000132083893},{"id":"https://openalex.org/C19024347","wikidata":"https://www.wikidata.org/wiki/Q211496","display_name":"High-level programming language","level":3,"score":0.3580999970436096},{"id":"https://openalex.org/C41681595","wikidata":"https://www.wikidata.org/wiki/Q7917855","display_name":"Vectorization (mathematics)","level":2,"score":0.35659998655319214},{"id":"https://openalex.org/C12186640","wikidata":"https://www.wikidata.org/wiki/Q6815743","display_name":"Memory model","level":3,"score":0.3492000102996826},{"id":"https://openalex.org/C47487241","wikidata":"https://www.wikidata.org/wiki/Q5227230","display_name":"Data access","level":2,"score":0.3409999907016754},{"id":"https://openalex.org/C190902152","wikidata":"https://www.wikidata.org/wiki/Q1325106","display_name":"Optimizing compiler","level":3,"score":0.3285999894142151},{"id":"https://openalex.org/C2776257435","wikidata":"https://www.wikidata.org/wiki/Q1576430","display_name":"Bandwidth (computing)","level":2,"score":0.32260000705718994},{"id":"https://openalex.org/C8767382","wikidata":"https://www.wikidata.org/wiki/Q1058454","display_name":"Dynamic compilation","level":3,"score":0.32170000672340393},{"id":"https://openalex.org/C162319229","wikidata":"https://www.wikidata.org/wiki/Q175263","display_name":"Data structure","level":2,"score":0.32019999623298645},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.3165999948978424},{"id":"https://openalex.org/C118524514","wikidata":"https://www.wikidata.org/wiki/Q173212","display_name":"Computer architecture","level":1,"score":0.31450000405311584},{"id":"https://openalex.org/C202491316","wikidata":"https://www.wikidata.org/wiki/Q272683","display_name":"Instruction set","level":2,"score":0.2953999936580658},{"id":"https://openalex.org/C35390924","wikidata":"https://www.wikidata.org/wiki/Q661075","display_name":"Metaprogramming","level":2,"score":0.29350000619888306},{"id":"https://openalex.org/C2776834041","wikidata":"https://www.wikidata.org/wiki/Q25346349","display_name":"Execution model","level":2,"score":0.28700000047683716},{"id":"https://openalex.org/C138958017","wikidata":"https://www.wikidata.org/wiki/Q190087","display_name":"Data type","level":2,"score":0.2759999930858612},{"id":"https://openalex.org/C150552126","wikidata":"https://www.wikidata.org/wiki/Q339387","display_name":"SIMD","level":2,"score":0.2662000060081482},{"id":"https://openalex.org/C83283714","wikidata":"https://www.wikidata.org/wiki/Q121117","display_name":"Supercomputer","level":2,"score":0.2660999894142151},{"id":"https://openalex.org/C133875982","wikidata":"https://www.wikidata.org/wiki/Q764810","display_name":"Shared memory","level":2,"score":0.2639999985694885}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3760250.3762219","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3760250.3762219","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3760250.3762219","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 31st ACM International Conference on Architectural Support for Programming Languages and Operating Systems, Volume 1","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.1145/3760250.3762219","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3760250.3762219","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3760250.3762219","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 31st ACM International Conference on Architectural Support for Programming Languages and Operating Systems, Volume 1","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G1597412403","display_name":null,"funder_award_id":"RGPIN-","funder_id":"https://openalex.org/F4320334593","funder_display_name":"Natural Sciences and Engineering Research Council of Canada"},{"id":"https://openalex.org/G2165548363","display_name":null,"funder_award_id":"Canada","funder_id":"https://openalex.org/F4320334593","funder_display_name":"Natural Sciences and Engineering Research Council of Canada"},{"id":"https://openalex.org/G2998137668","display_name":null,"funder_award_id":"unknown","funder_id":"https://openalex.org/F4320334593","funder_display_name":"Natural Sciences and Engineering Research Council of Canada"},{"id":"https://openalex.org/G3033334534","display_name":null,"funder_award_id":"AI Chair","funder_id":"https://openalex.org/F4320309949","funder_display_name":"Canadian Institute for Advanced Research"},{"id":"https://openalex.org/G400116904","display_name":null,"funder_award_id":"unknow","funder_id":"https://openalex.org/F4320334593","funder_display_name":"Natural Sciences and Engineering Research Council of Canada"},{"id":"https://openalex.org/G5784215521","display_name":null,"funder_award_id":"Chair","funder_id":"https://openalex.org/F4320334593","funder_display_name":"Natural Sciences and Engineering Research Council of Canada"},{"id":"https://openalex.org/G6221715925","display_name":null,"funder_award_id":"RGPIN","funder_id":"https://openalex.org/F4320334593","funder_display_name":"Natural Sciences and Engineering Research Council of Canada"},{"id":"https://openalex.org/G8105784103","display_name":null,"funder_award_id":"RGPIN-202","funder_id":"https://openalex.org/F4320334593","funder_display_name":"Natural Sciences and Engineering Research Council of Canada"}],"funders":[{"id":"https://openalex.org/F4320309949","display_name":"Canadian Institute for Advanced Research","ror":"https://ror.org/01sdtdd95"},{"id":"https://openalex.org/F4320316785","display_name":"VMware","ror":null},{"id":"https://openalex.org/F4320322015","display_name":"University of Toronto","ror":"https://ror.org/03dbr7087"},{"id":"https://openalex.org/F4320334593","display_name":"Natural Sciences and Engineering Research Council of Canada","ror":"https://ror.org/01h531d29"}],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4417241966.pdf","grobid_xml":"https://content.openalex.org/works/W4417241966.grobid-xml"},"referenced_works_count":23,"referenced_works":["https://openalex.org/W2000335122","https://openalex.org/W2954698171","https://openalex.org/W2961619211","https://openalex.org/W2970419734","https://openalex.org/W2979310060","https://openalex.org/W2981758446","https://openalex.org/W3012249773","https://openalex.org/W3161395920","https://openalex.org/W3205226039","https://openalex.org/W4212986322","https://openalex.org/W4233996382","https://openalex.org/W4251637954","https://openalex.org/W4281710230","https://openalex.org/W4281932226","https://openalex.org/W4318541538","https://openalex.org/W4318541647","https://openalex.org/W4321446237","https://openalex.org/W4327930469","https://openalex.org/W4387321091","https://openalex.org/W4404955120","https://openalex.org/W4407196927","https://openalex.org/W4408151434","https://openalex.org/W4409362928"],"related_works":[],"abstract_inverted_index":{"Serving":[0],"Large":[1],"Language":[2],"Models":[3],"(LLMs)":[4],"is":[5],"critical":[6,69],"for":[7,40,84,98,138],"AI-powered":[8],"applications,":[9],"yet":[10],"it":[11],"demands":[12],"substantial":[13],"computational":[14,21],"resources,":[15],"particularly":[16],"in":[17],"memory":[18,78,128],"bandwidth":[19],"and":[20,55,76,135,155,172,184,194,204],"throughput.":[22],"Low-precision":[23],"computation":[24],"has":[25],"emerged":[26],"as":[27,72,182,186,188,192],"a":[28,94,122,126,130,165],"key":[29],"technique":[30],"to":[31,46,114,178],"improve":[32],"efficiency":[33],"while":[34,116],"reducing":[35],"resource":[36],"consumption.":[37],"Existing":[38],"approaches":[39],"generating":[41],"low-precision":[42,86,105,140,169,175],"kernels":[43,190],"are":[44,51,82,145],"limited":[45],"weight":[47],"bit":[48,110],"widths":[49,111],"that":[50,81,103,161],"powers":[52],"of":[53,61,168],"two":[54],"suffer":[56],"from":[57,112],"suboptimal":[58],"performance":[59,198],"because":[60],"high-level":[62],"GPU":[63,100,118,150],"programming":[64,124],"abstractions.":[65],"These":[66],"abstractions":[67],"restrict":[68],"optimizations,":[70],"such":[71,181,191],"fine-grained":[73],"register":[74],"management":[75],"optimized":[77],"access":[79],"patterns,":[80],"essential":[83],"efficient":[85,149],"computations.":[87],"In":[88],"this":[89],"paper,":[90],"we":[91],"introduce":[92],"Tilus,":[93],"domain-specific":[95],"language":[96],"designed":[97],"General-Purpose":[99],"(GPGPU)":[101],"computing":[102],"supports":[104,164],"data":[106,141,170],"types":[107],"with":[108],"arbitrary":[109],"1":[113],"8":[115],"maintaining":[117],"programmability.":[119],"Tilus":[120,143,162,196,209],"features":[121],"thread-block-level":[123],"model,":[125],"hierarchical":[127],"space,":[129],"novel":[131],"algebraic":[132],"layout":[133],"system,":[134],"extensive":[136],"support":[137],"diverse":[139],"types.":[142],"programs":[144,151],"compiled":[146],"into":[147],"highly":[148],"through":[152],"automatic":[153],"vectorization":[154],"instruction":[156],"selection.":[157],"Extensive":[158],"experiments":[159],"demonstrate":[160],"efficiently":[163],"full":[166],"spectrum":[167],"types,":[171],"outperforms":[173],"state-of-the-art":[174],"kernels.":[176],"Compared":[177],"existing":[179],"compilers":[180],"Triton":[183],"Ladder,":[185],"well":[187],"hand-optimized":[189],"QuantLLM":[193],"Marlin,":[195],"achieves":[197],"improvements":[199],"of:":[200],"1.75x,":[201],"2.61x,":[202],"1.29x":[203],"1.03x,":[205],"respectively.":[206],"We":[207],"open-source":[208],"at":[210],"https://github.com/NVIDIA/tilus.":[211]},"counts_by_year":[],"updated_date":"2026-03-27T05:58:40.876381","created_date":"2025-12-11T00:00:00"}