{"id":"https://openalex.org/W7139095235","doi":"https://doi.org/10.1109/tpds.2026.3675079","title":"Texplorer: Efficient Tensor Program Optimization for GPUs Using a Highly Constrained Search Space","display_name":"Texplorer: Efficient Tensor Program Optimization for GPUs Using a Highly Constrained Search Space","publication_year":2026,"publication_date":"2026-03-18","ids":{"openalex":"https://openalex.org/W7139095235","doi":"https://doi.org/10.1109/tpds.2026.3675079"},"language":null,"primary_location":{"id":"doi:10.1109/tpds.2026.3675079","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tpds.2026.3675079","pdf_url":null,"source":{"id":"https://openalex.org/S97130795","display_name":"IEEE Transactions on Parallel and Distributed Systems","issn_l":"1045-9219","issn":["1045-9219","1558-2183","2161-9883"],"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 Parallel and Distributed Systems","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":null,"display_name":"Wei Li","orcid":"https://orcid.org/0000-0002-9763-3319"},"institutions":[{"id":"https://openalex.org/I165932596","display_name":"National University of Singapore","ror":"https://ror.org/01tgyzw49","country_code":"SG","type":"education","lineage":["https://openalex.org/I165932596"]}],"countries":["SG"],"is_corresponding":false,"raw_author_name":"Wei Li","raw_affiliation_strings":["School of Computing, National University of Singapore, Singapore"],"raw_orcid":"https://orcid.org/0000-0002-9763-3319","affiliations":[{"raw_affiliation_string":"School of Computing, National University of Singapore, Singapore","institution_ids":["https://openalex.org/I165932596"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Bin Gao","orcid":"https://orcid.org/0000-0001-5009-3514"},"institutions":[{"id":"https://openalex.org/I165932596","display_name":"National University of Singapore","ror":"https://ror.org/01tgyzw49","country_code":"SG","type":"education","lineage":["https://openalex.org/I165932596"]}],"countries":["SG"],"is_corresponding":false,"raw_author_name":"Bin Gao","raw_affiliation_strings":["School of Computing, National University of Singapore, Singapore"],"raw_orcid":"https://orcid.org/0000-0001-5009-3514","affiliations":[{"raw_affiliation_string":"School of Computing, National University of Singapore, Singapore","institution_ids":["https://openalex.org/I165932596"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5129968893","display_name":"Weng-Fai Wong","orcid":null},"institutions":[{"id":"https://openalex.org/I165932596","display_name":"National University of Singapore","ror":"https://ror.org/01tgyzw49","country_code":"SG","type":"education","lineage":["https://openalex.org/I165932596"]}],"countries":["SG"],"is_corresponding":false,"raw_author_name":"Weng-Fai Wong","raw_affiliation_strings":["School of Computing, National University of Singapore, Singapore"],"raw_orcid":"https://orcid.org/0000-0002-4281-2053","affiliations":[{"raw_affiliation_string":"School of Computing, National University of Singapore, Singapore","institution_ids":["https://openalex.org/I165932596"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I165932596"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.28494603,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":"37","issue":"5","first_page":"1205","last_page":"1218"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11612","display_name":"Stochastic Gradient Optimization Techniques","score":0.18199999630451202,"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/T11612","display_name":"Stochastic Gradient Optimization Techniques","score":0.18199999630451202,"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":0.15629999339580536,"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/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.11630000174045563,"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/pruning","display_name":"Pruning","score":0.7667999863624573},{"id":"https://openalex.org/keywords/pipeline","display_name":"Pipeline (software)","score":0.626800000667572},{"id":"https://openalex.org/keywords/tensor","display_name":"Tensor (intrinsic definition)","score":0.4970000088214874},{"id":"https://openalex.org/keywords/key","display_name":"Key (lock)","score":0.4772999882698059},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.4496000111103058},{"id":"https://openalex.org/keywords/convergence","display_name":"Convergence (economics)","score":0.4302999973297119},{"id":"https://openalex.org/keywords/search-algorithm","display_name":"Search algorithm","score":0.4300999939441681},{"id":"https://openalex.org/keywords/beam-search","display_name":"Beam search","score":0.4153999984264374},{"id":"https://openalex.org/keywords/bounding-overwatch","display_name":"Bounding overwatch","score":0.37310001254081726},{"id":"https://openalex.org/keywords/beam-stack-search","display_name":"Beam stack search","score":0.36480000615119934}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8456000089645386},{"id":"https://openalex.org/C108010975","wikidata":"https://www.wikidata.org/wiki/Q500094","display_name":"Pruning","level":2,"score":0.7667999863624573},{"id":"https://openalex.org/C43521106","wikidata":"https://www.wikidata.org/wiki/Q2165493","display_name":"Pipeline (software)","level":2,"score":0.626800000667572},{"id":"https://openalex.org/C155281189","wikidata":"https://www.wikidata.org/wiki/Q3518150","display_name":"Tensor (intrinsic definition)","level":2,"score":0.4970000088214874},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.4772999882698059},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.4496000111103058},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.43140000104904175},{"id":"https://openalex.org/C2777303404","wikidata":"https://www.wikidata.org/wiki/Q759757","display_name":"Convergence (economics)","level":2,"score":0.4302999973297119},{"id":"https://openalex.org/C125583679","wikidata":"https://www.wikidata.org/wiki/Q755673","display_name":"Search algorithm","level":2,"score":0.4300999939441681},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.41659998893737793},{"id":"https://openalex.org/C19889080","wikidata":"https://www.wikidata.org/wiki/Q2835852","display_name":"Beam search","level":3,"score":0.4153999984264374},{"id":"https://openalex.org/C63584917","wikidata":"https://www.wikidata.org/wiki/Q333286","display_name":"Bounding overwatch","level":2,"score":0.37310001254081726},{"id":"https://openalex.org/C194036150","wikidata":"https://www.wikidata.org/wiki/Q4876196","display_name":"Beam stack search","level":5,"score":0.36480000615119934},{"id":"https://openalex.org/C21782646","wikidata":"https://www.wikidata.org/wiki/Q841666","display_name":"Search cost","level":2,"score":0.36410000920295715},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.33410000801086426},{"id":"https://openalex.org/C165696696","wikidata":"https://www.wikidata.org/wiki/Q11287","display_name":"Exploit","level":2,"score":0.3310999870300293},{"id":"https://openalex.org/C165064840","wikidata":"https://www.wikidata.org/wiki/Q1321061","display_name":"Matching (statistics)","level":2,"score":0.3301999866962433},{"id":"https://openalex.org/C201789804","wikidata":"https://www.wikidata.org/wiki/Q2362762","display_name":"Search problem","level":2,"score":0.32690000534057617},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.32199999690055847},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.32120001316070557},{"id":"https://openalex.org/C2778572836","wikidata":"https://www.wikidata.org/wiki/Q380933","display_name":"Space (punctuation)","level":2,"score":0.30070000886917114},{"id":"https://openalex.org/C97854310","wikidata":"https://www.wikidata.org/wiki/Q19541","display_name":"Search engine","level":2,"score":0.2976999878883362},{"id":"https://openalex.org/C203208320","wikidata":"https://www.wikidata.org/wiki/Q5150830","display_name":"Combinatorial search","level":4,"score":0.2921000123023987},{"id":"https://openalex.org/C189430467","wikidata":"https://www.wikidata.org/wiki/Q7293293","display_name":"Ranking (information retrieval)","level":2,"score":0.287200003862381},{"id":"https://openalex.org/C74912251","wikidata":"https://www.wikidata.org/wiki/Q6815727","display_name":"Memory footprint","level":2,"score":0.2851000130176544},{"id":"https://openalex.org/C135320971","wikidata":"https://www.wikidata.org/wiki/Q1868524","display_name":"Local search (optimization)","level":2,"score":0.2741999924182892},{"id":"https://openalex.org/C14362708","wikidata":"https://www.wikidata.org/wiki/Q1675274","display_name":"Iterative deepening depth-first search","level":5,"score":0.2718999981880188},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.26660001277923584},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.26510000228881836},{"id":"https://openalex.org/C137836250","wikidata":"https://www.wikidata.org/wiki/Q984063","display_name":"Optimization problem","level":2,"score":0.26429998874664307},{"id":"https://openalex.org/C157486923","wikidata":"https://www.wikidata.org/wiki/Q1376436","display_name":"String (physics)","level":2,"score":0.262800008058548},{"id":"https://openalex.org/C2778100165","wikidata":"https://www.wikidata.org/wiki/Q1589327","display_name":"Memory hierarchy","level":3,"score":0.2603999972343445},{"id":"https://openalex.org/C2778119891","wikidata":"https://www.wikidata.org/wiki/Q477690","display_name":"CUDA","level":2,"score":0.25589999556541443},{"id":"https://openalex.org/C159694833","wikidata":"https://www.wikidata.org/wiki/Q2321565","display_name":"Iterative method","level":2,"score":0.2533000111579895},{"id":"https://openalex.org/C50630238","wikidata":"https://www.wikidata.org/wiki/Q971505","display_name":"General-purpose computing on graphics processing units","level":3,"score":0.2531999945640564}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/tpds.2026.3675079","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tpds.2026.3675079","pdf_url":null,"source":{"id":"https://openalex.org/S97130795","display_name":"IEEE Transactions on Parallel and Distributed Systems","issn_l":"1045-9219","issn":["1045-9219","1558-2183","2161-9883"],"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 Parallel and Distributed Systems","raw_type":"journal-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[{"id":"https://openalex.org/F4320320751","display_name":"Ministry of Education - Singapore","ror":"https://ror.org/01kcva023"}],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Optimizing":[0],"tensor":[1,63,79],"programs":[2,38,64],"for":[3,86],"deep":[4],"learning":[5],"inference":[6],"is":[7,50],"costly,":[8],"often":[9],"requiring":[10],"hours":[11],"of":[12,56],"search":[13,16,36,48,106,185],"over":[14],"massive":[15],"spaces":[17],"(<inline-formula":[18],"xmlns:mml=\"http://www.w3.org/1998/Math/MathML\"":[19,191,198,212],"xmlns:xlink=\"http://www.w3.org/1999/xlink\"><tex-math":[20,192,213],"notation=\"LaTeX\">$\\gt":[21],"10^{10}$</tex-math></inline-formula>).":[22],"Prior":[23],"machine":[24],"learning-guided":[25],"approaches":[26],"are":[27],"stochastic":[28,164],"and":[29,37,61,71,88,101,127,135,143,169,176],"inefficient":[30,102],"because":[31],"they":[32],"treat":[33],"both":[34],"the":[35,47,105],"as":[39],"black":[40],"boxes,":[41],"overlooking":[42],"two":[43],"key":[44],"properties:":[45],"(1)":[46],"space":[49,107],"highly":[51],"structured,":[52],"with":[53,117,208],"large":[54],"regions":[55],"invalid":[57],"or":[58,163],"low-performing":[59],"configurations,":[60],"(2)":[62],"exhibit":[65],"analyzable":[66],"patterns":[67],"in":[68],"memory":[69,128],"access":[70],"pipelined":[72],"execution.":[73],"We":[74],"present":[75],"Texplorer,":[76],"a":[77,93,118,147],"deterministic":[78],"program":[80],"optimizer":[81],"that":[82,123,152],"exploits":[83],"these":[84],"structures":[85],"robust":[87],"efficient":[89],"search.":[90],"Texplorer":[91,140,179],"introduces":[92],"three-stage":[94],"pruning":[95,142],"algorithm":[96],"to":[97,110,189,206,210],"eliminate":[98],"invalid,":[99],"redundant,":[100],"candidates,":[103,156],"shrinking":[104],"from":[108],"billions":[109],"thousands.":[111],"It":[112],"replaces":[113],"costly":[114],"learned":[115],"models":[116],"lightweight":[119],"analytical":[120,149],"cost":[121,144],"model":[122],"captures":[124],"GPU":[125],"pipeline":[126,151],"behavior,":[129],"enabling":[130,157],"accurate,":[131],"training-free":[132],"performance":[133,182,203],"prediction":[134],"strong":[136],"cross-operator":[137],"generalization.":[138],"Finally,":[139],"integrates":[141],"modeling":[145],"into":[146],"one-pass":[148],"evaluation":[150],"ranks":[153],"all":[154],"remaining":[155],"rapid":[158],"convergence":[159],"without":[160],"iterative":[161],"retraining":[162],"sampling.":[165],"Across":[166],"121":[167],"operators":[168],"11":[170],"DNNs":[171],"on":[172],"NVIDIA":[173],"V100,":[174],"A100,":[175],"H100":[177],"GPUs,":[178],"matches":[180],"state-of-the-art":[181],"while":[183],"reducing":[184],"time":[186],"by":[187,204],"up":[188,205,209],"<inline-formula":[190,211],"notation=\"LaTeX\">$300\\times$</tex-math></inline-formula>.":[193],"An":[194],"enhanced":[195],"variant,":[196],"<bold":[197],"xmlns:xlink=\"http://www.w3.org/1999/xlink\">Texplorer<inline-formula><tex-math":[199],"notation=\"LaTeX\">$^+$</tex-math></inline-formula></b>,":[200],"further":[201],"improves":[202],"3.1%":[207],"notation=\"LaTeX\">$72\\times$</tex-math></inline-formula>":[214],"faster":[215],"tuning.":[216]},"counts_by_year":[],"updated_date":"2026-04-07T06:01:17.266235","created_date":"2026-03-20T00:00:00"}