{"id":"https://openalex.org/W7131131434","doi":"https://doi.org/10.1109/cgo68049.2026.11394842","title":"Fast Autoscheduling for Sparse ML Frameworks","display_name":"Fast Autoscheduling for Sparse ML Frameworks","publication_year":2026,"publication_date":"2026-01-31","ids":{"openalex":"https://openalex.org/W7131131434","doi":"https://doi.org/10.1109/cgo68049.2026.11394842"},"language":null,"primary_location":{"id":"doi:10.1109/cgo68049.2026.11394842","is_oa":false,"landing_page_url":"https://doi.org/10.1109/cgo68049.2026.11394842","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2026 IEEE/ACM International Symposium on Code Generation and Optimization (CGO)","raw_type":"proceedings-article"},"type":"conference-paper","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":"https://openalex.org/A5019401621","display_name":"Bobby Yan","orcid":"https://orcid.org/0009-0002-6792-6222"},"institutions":[{"id":"https://openalex.org/I97018004","display_name":"Stanford University","ror":"https://ror.org/00f54p054","country_code":"US","type":"education","lineage":["https://openalex.org/I97018004"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Bobby Yan","raw_affiliation_strings":["Stanford University,Stanford,USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Stanford University,Stanford,USA","institution_ids":["https://openalex.org/I97018004"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5034921995","display_name":"Alexander J Root","orcid":"https://orcid.org/0000-0001-6221-1389"},"institutions":[{"id":"https://openalex.org/I97018004","display_name":"Stanford University","ror":"https://ror.org/00f54p054","country_code":"US","type":"education","lineage":["https://openalex.org/I97018004"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Alexander J Root","raw_affiliation_strings":["Stanford University,Stanford,USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Stanford University,Stanford,USA","institution_ids":["https://openalex.org/I97018004"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5032764769","display_name":"Trevor Gale","orcid":"https://orcid.org/0000-0003-3927-9267"},"institutions":[{"id":"https://openalex.org/I97018004","display_name":"Stanford University","ror":"https://ror.org/00f54p054","country_code":"US","type":"education","lineage":["https://openalex.org/I97018004"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Trevor Gale","raw_affiliation_strings":["Stanford University,Stanford,USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Stanford University,Stanford,USA","institution_ids":["https://openalex.org/I97018004"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5073873489","display_name":"David Broman","orcid":"https://orcid.org/0000-0001-8457-4105"},"institutions":[{"id":"https://openalex.org/I86987016","display_name":"KTH Royal Institute of Technology","ror":"https://ror.org/026vcq606","country_code":"SE","type":"education","lineage":["https://openalex.org/I86987016"]}],"countries":["SE"],"is_corresponding":false,"raw_author_name":"David Broman","raw_affiliation_strings":["KTH,Stockholm,Sweden"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"KTH,Stockholm,Sweden","institution_ids":["https://openalex.org/I86987016"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5041886781","display_name":"Fredrik Kj\u00f8lstad","orcid":"https://orcid.org/0000-0002-2267-903X"},"institutions":[{"id":"https://openalex.org/I97018004","display_name":"Stanford University","ror":"https://ror.org/00f54p054","country_code":"US","type":"education","lineage":["https://openalex.org/I97018004"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Fredrik Kjolstad","raw_affiliation_strings":["Stanford University,Stanford,USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Stanford University,Stanford,USA","institution_ids":["https://openalex.org/I97018004"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.11828442,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"28","last_page":"43"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.34220001101493835,"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":0.34220001101493835,"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/T10904","display_name":"Embedded Systems Design Techniques","score":0.1696999967098236,"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/T12535","display_name":"Machine Learning and Data Classification","score":0.08209999650716782,"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/inference","display_name":"Inference","score":0.555400013923645},{"id":"https://openalex.org/keywords/computation","display_name":"Computation","score":0.5550000071525574},{"id":"https://openalex.org/keywords/sparse-matrix","display_name":"Sparse matrix","score":0.5357999801635742},{"id":"https://openalex.org/keywords/neural-coding","display_name":"Neural coding","score":0.48420000076293945},{"id":"https://openalex.org/keywords/tensor","display_name":"Tensor (intrinsic definition)","score":0.4546000063419342},{"id":"https://openalex.org/keywords/compiler","display_name":"Compiler","score":0.430400013923645},{"id":"https://openalex.org/keywords/deep-learning","display_name":"Deep learning","score":0.42910000681877136},{"id":"https://openalex.org/keywords/sparse-approximation","display_name":"Sparse approximation","score":0.37299999594688416},{"id":"https://openalex.org/keywords/scheduling","display_name":"Scheduling (production processes)","score":0.3614000082015991}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7802000045776367},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.555400013923645},{"id":"https://openalex.org/C45374587","wikidata":"https://www.wikidata.org/wiki/Q12525525","display_name":"Computation","level":2,"score":0.5550000071525574},{"id":"https://openalex.org/C56372850","wikidata":"https://www.wikidata.org/wiki/Q1050404","display_name":"Sparse matrix","level":3,"score":0.5357999801635742},{"id":"https://openalex.org/C77637269","wikidata":"https://www.wikidata.org/wiki/Q7002051","display_name":"Neural coding","level":2,"score":0.48420000076293945},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.47350001335144043},{"id":"https://openalex.org/C155281189","wikidata":"https://www.wikidata.org/wiki/Q3518150","display_name":"Tensor (intrinsic definition)","level":2,"score":0.4546000063419342},{"id":"https://openalex.org/C169590947","wikidata":"https://www.wikidata.org/wiki/Q47506","display_name":"Compiler","level":2,"score":0.430400013923645},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.42910000681877136},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.39660000801086426},{"id":"https://openalex.org/C124066611","wikidata":"https://www.wikidata.org/wiki/Q28684319","display_name":"Sparse approximation","level":2,"score":0.37299999594688416},{"id":"https://openalex.org/C206729178","wikidata":"https://www.wikidata.org/wiki/Q2271896","display_name":"Scheduling (production processes)","level":2,"score":0.3614000082015991},{"id":"https://openalex.org/C43126263","wikidata":"https://www.wikidata.org/wiki/Q128751","display_name":"Source code","level":2,"score":0.3605000078678131},{"id":"https://openalex.org/C13251829","wikidata":"https://www.wikidata.org/wiki/Q3085841","display_name":"Dense graph","level":5,"score":0.35899999737739563},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3555000126361847},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.33009999990463257},{"id":"https://openalex.org/C132525143","wikidata":"https://www.wikidata.org/wiki/Q141488","display_name":"Graph","level":2,"score":0.3203999996185303},{"id":"https://openalex.org/C2986737658","wikidata":"https://www.wikidata.org/wiki/Q30103009","display_name":"Tensor decomposition","level":3,"score":0.3188999891281128},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.3165999948978424},{"id":"https://openalex.org/C83283714","wikidata":"https://www.wikidata.org/wiki/Q121117","display_name":"Supercomputer","level":2,"score":0.29750001430511475},{"id":"https://openalex.org/C63479239","wikidata":"https://www.wikidata.org/wiki/Q7353546","display_name":"Robustness (evolution)","level":3,"score":0.29499998688697815},{"id":"https://openalex.org/C2777472644","wikidata":"https://www.wikidata.org/wiki/Q16968992","display_name":"Approximate inference","level":3,"score":0.29350000619888306},{"id":"https://openalex.org/C162319229","wikidata":"https://www.wikidata.org/wiki/Q175263","display_name":"Data structure","level":2,"score":0.2874000072479248},{"id":"https://openalex.org/C2776760102","wikidata":"https://www.wikidata.org/wiki/Q5139990","display_name":"Code (set theory)","level":3,"score":0.28369998931884766},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.2800999879837036},{"id":"https://openalex.org/C137836250","wikidata":"https://www.wikidata.org/wiki/Q984063","display_name":"Optimization problem","level":2,"score":0.2628999948501587},{"id":"https://openalex.org/C190902152","wikidata":"https://www.wikidata.org/wiki/Q1325106","display_name":"Optimizing compiler","level":3,"score":0.2547000050544739}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/cgo68049.2026.11394842","is_oa":false,"landing_page_url":"https://doi.org/10.1109/cgo68049.2026.11394842","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2026 IEEE/ACM International Symposium on Code Generation and Optimization (CGO)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/9","display_name":"Industry, innovation and infrastructure","score":0.42215102910995483}],"awards":[],"funders":[{"id":"https://openalex.org/F4320306076","display_name":"National Science Foundation","ror":"https://ror.org/021nxhr62"}],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":33,"referenced_works":["https://openalex.org/W1834627138","https://openalex.org/W2009654791","https://openalex.org/W2035080386","https://openalex.org/W2055312318","https://openalex.org/W2112796928","https://openalex.org/W2119609467","https://openalex.org/W2126004407","https://openalex.org/W2153959628","https://openalex.org/W2162630660","https://openalex.org/W2168190036","https://openalex.org/W2590246587","https://openalex.org/W2808133870","https://openalex.org/W2898123186","https://openalex.org/W2902783593","https://openalex.org/W2949967139","https://openalex.org/W2961619211","https://openalex.org/W3028129830","https://openalex.org/W3108012228","https://openalex.org/W3112890400","https://openalex.org/W3121828480","https://openalex.org/W3130660608","https://openalex.org/W4220690649","https://openalex.org/W4255450819","https://openalex.org/W4281658036","https://openalex.org/W4290648346","https://openalex.org/W4321500415","https://openalex.org/W4327911434","https://openalex.org/W4379518528","https://openalex.org/W4379537250","https://openalex.org/W4394998532","https://openalex.org/W4399852843","https://openalex.org/W4403223049","https://openalex.org/W4403223519"],"related_works":[],"abstract_inverted_index":{"The":[0],"rapid":[1],"growth":[2],"in":[3,99,149,169],"the":[4,11,62,150],"size":[5],"of":[6,13,80,153],"deep":[7,32],"learning":[8,33,68],"models":[9],"strains":[10],"capabilities":[12],"dense":[14],"computation":[15,19],"paradigms.":[16],"Leveraging":[17],"sparse":[18,39,81,133,154,173,203,206],"has":[20],"become":[21],"increasingly":[22],"popular":[23],"for":[24,38,111,136,176,213],"training":[25],"and":[26,138,160,205],"deploying":[27],"large-scale":[28],"models,":[29],"but":[30],"existing":[31],"frameworks":[34],"lack":[35],"extensive":[36],"support":[37],"operations.":[40],"Current":[41],"approaches":[42],"either":[43],"require":[44],"manual":[45],"scheduling":[46],"expertise":[47],"or":[48],"rely":[49],"on":[50,142,196],"exhaustive":[51],"search":[52],"taking":[53],"hours":[54],"to":[55,66,119],"days,":[56],"which":[57],"are":[58,147],"both":[59],"incompatible":[60],"with":[61,208],"interactive":[63,214],"development":[64],"essential":[65],"machine":[67],"research.":[69],"We":[70,165],"present":[71,124],"three":[72],"algorithmic":[73],"contributions":[74],"that":[75,92,115,129,178],"enable":[76],"fast,":[77],"automatic":[78],"optimization":[79],"tensor":[82,134,155,174],"computations.":[83],"First,":[84],"we":[85,105,123],"develop":[86],"a":[87,107,125,171],"heuristic-based":[88],"loop":[89],"ordering":[90],"algorithm":[91,109,128],"avoids":[93],"asymptotic":[94],"performance":[95,117],"cliffs":[96],"while":[97],"compiling":[98],"milliseconds":[100],"rather":[101],"than":[102],"hours.":[103],"Second,":[104],"introduce":[106],"tiling":[108],"specialized":[110],"mixed":[112],"sparse-dense":[113],"computations":[114],"achieves":[116,190],"comparable":[118],"hand-optimized":[120],"kernels.":[121],"Third,":[122],"format":[126],"inference":[127],"automatically":[130],"selects":[131],"appropriate":[132],"formats":[135],"intermediate":[137],"output":[139],"tensors":[140],"based":[141],"operation":[143],"semantics.":[144],"These":[145],"algorithms":[146],"grounded":[148],"computational":[151],"properties":[152],"algebra,":[156],"making":[157],"them":[158],"predictable":[159],"robust":[161],"across":[162],"diverse":[163],"workloads.":[164],"implement":[166],"these":[167],"techniques":[168],"Scorch,":[170],"prototype":[172],"compiler":[175],"PyTorch":[177,194],"demonstrates":[179],"their":[180],"practical":[181],"effectiveness.":[182],"With":[183],"only":[184],"minimal":[185],"code":[186],"changes,":[187],"our":[188],"approach":[189],"1.05\u20135.80\u00d7":[191],"speedups":[192],"over":[193],"Sparse":[195],"end-to-end":[197],"tasks":[198],"including":[199],"graph":[200],"neural":[201],"networks,":[202],"autoencoders,":[204],"transformers,":[207],"compilation":[209],"times":[210],"fast":[211],"enough":[212],"ML":[215],"development.":[216]},"counts_by_year":[],"updated_date":"2026-07-29T14:22:42.915294","created_date":"2026-02-24T00:00:00"}