{"id":"https://openalex.org/W4366342693","doi":"https://doi.org/10.1145/3815424","title":"STen: Productive and Efficient Sparsity in PyTorch","display_name":"STen: Productive and Efficient Sparsity in PyTorch","publication_year":2026,"publication_date":"2026-05-27","ids":{"openalex":"https://openalex.org/W4366342693","doi":"https://doi.org/10.1145/3815424"},"language":"en","primary_location":{"id":"doi:10.1145/3815424","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3815424","pdf_url":null,"source":{"id":"https://openalex.org/S26056741","display_name":"ACM Transactions on Architecture and Code Optimization","issn_l":"1544-3566","issn":["1544-3566","1544-3973"],"is_oa":true,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319798","host_organization_name":"Association for Computing Machinery","host_organization_lineage":["https://openalex.org/P4310319798"],"host_organization_lineage_names":["Association for Computing Machinery"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ACM Transactions on Architecture and Code Optimization","raw_type":"journal-article"},"type":"article","indexed_in":["arxiv","crossref","datacite"],"open_access":{"is_oa":true,"oa_status":"diamond","oa_url":"https://doi.org/10.1145/3815424","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5100699645","display_name":"\u0410\u043d\u0434\u0440\u0435\u0439 \u0418\u0432\u0430\u043d\u043e\u0432","orcid":"https://orcid.org/0000-0003-1888-5612"},"institutions":[{"id":"https://openalex.org/I35440088","display_name":"ETH Zurich","ror":"https://ror.org/05a28rw58","country_code":"CH","type":"education","lineage":["https://openalex.org/I2799323385","https://openalex.org/I35440088"]}],"countries":["CH"],"is_corresponding":false,"raw_author_name":"Andrei Ivanov","raw_affiliation_strings":["ETH Zurich"],"raw_orcid":"https://orcid.org/0009-0007-9487-9990","affiliations":[{"raw_affiliation_string":"ETH Zurich","institution_ids":["https://openalex.org/I35440088"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5066306669","display_name":"Nikoli Dryden","orcid":null},"institutions":[{"id":"https://openalex.org/I35440088","display_name":"ETH Zurich","ror":"https://ror.org/05a28rw58","country_code":"CH","type":"education","lineage":["https://openalex.org/I2799323385","https://openalex.org/I35440088"]}],"countries":["CH"],"is_corresponding":false,"raw_author_name":"Nikoli Dryden","raw_affiliation_strings":["ETH Zurich"],"raw_orcid":"https://orcid.org/0000-0002-9965-3647","affiliations":[{"raw_affiliation_string":"ETH Zurich","institution_ids":["https://openalex.org/I35440088"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5002615744","display_name":"Tal Ben\u2010Nun","orcid":"https://orcid.org/0000-0002-3657-6568"},"institutions":[{"id":"https://openalex.org/I35440088","display_name":"ETH Zurich","ror":"https://ror.org/05a28rw58","country_code":"CH","type":"education","lineage":["https://openalex.org/I2799323385","https://openalex.org/I35440088"]}],"countries":["CH"],"is_corresponding":false,"raw_author_name":"Tal Ben-Nun","raw_affiliation_strings":["ETH Zurich"],"raw_orcid":"https://orcid.org/0000-0002-3657-6568","affiliations":[{"raw_affiliation_string":"ETH Zurich","institution_ids":["https://openalex.org/I35440088"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Timo Schneider","orcid":"https://orcid.org/0000-0002-4884-3934"},"institutions":[{"id":"https://openalex.org/I35440088","display_name":"ETH Zurich","ror":"https://ror.org/05a28rw58","country_code":"CH","type":"education","lineage":["https://openalex.org/I2799323385","https://openalex.org/I35440088"]}],"countries":["CH"],"is_corresponding":false,"raw_author_name":"Timo Schneider","raw_affiliation_strings":["ETH Zurich"],"raw_orcid":"https://orcid.org/0000-0002-4884-3934","affiliations":[{"raw_affiliation_string":"ETH Zurich","institution_ids":["https://openalex.org/I35440088"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5086409515","display_name":"Saleh Ashkboos","orcid":"https://orcid.org/0000-0001-6115-6779"},"institutions":[{"id":"https://openalex.org/I35440088","display_name":"ETH Zurich","ror":"https://ror.org/05a28rw58","country_code":"CH","type":"education","lineage":["https://openalex.org/I2799323385","https://openalex.org/I35440088"]}],"countries":["CH"],"is_corresponding":false,"raw_author_name":"Saleh Ashkboos","raw_affiliation_strings":["ETH Zurich"],"raw_orcid":"https://orcid.org/0000-0001-6115-6779","affiliations":[{"raw_affiliation_string":"ETH Zurich","institution_ids":["https://openalex.org/I35440088"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5026990786","display_name":"Torsten Hoefler","orcid":"https://orcid.org/0000-0002-1333-9797"},"institutions":[{"id":"https://openalex.org/I35440088","display_name":"ETH Zurich","ror":"https://ror.org/05a28rw58","country_code":"CH","type":"education","lineage":["https://openalex.org/I2799323385","https://openalex.org/I35440088"]}],"countries":["CH"],"is_corresponding":false,"raw_author_name":"Torsten Hoefler","raw_affiliation_strings":["ETH Zurich"],"raw_orcid":"https://orcid.org/0000-0002-1333-9797","affiliations":[{"raw_affiliation_string":"ETH Zurich","institution_ids":["https://openalex.org/I35440088"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I35440088"],"apc_list":null,"apc_paid":null,"fwci":7.365,"has_fulltext":true,"cited_by_count":2,"citation_normalized_percentile":{"value":0.92113455,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":90,"max":97},"biblio":{"volume":"23","issue":"2","first_page":"1","last_page":"25"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.9936000108718872,"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"}},"topics":[{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.9936000108718872,"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/T12303","display_name":"Tensor decomposition and applications","score":0.9901000261306763,"subfield":{"id":"https://openalex.org/subfields/2605","display_name":"Computational Mathematics"},"field":{"id":"https://openalex.org/fields/26","display_name":"Mathematics"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.980400025844574,"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/computer-science","display_name":"Computer science","score":0.7608030438423157},{"id":"https://openalex.org/keywords/pipeline","display_name":"Pipeline (software)","score":0.7069889903068542},{"id":"https://openalex.org/keywords/focus","display_name":"Focus (optics)","score":0.7049645185470581},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.630073606967926},{"id":"https://openalex.org/keywords/component","display_name":"Component (thermodynamics)","score":0.6056195497512817},{"id":"https://openalex.org/keywords/deep-learning","display_name":"Deep learning","score":0.5670874118804932},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.4663008749485016},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.45264485478401184},{"id":"https://openalex.org/keywords/tensor","display_name":"Tensor (intrinsic definition)","score":0.4414559602737427},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.36550402641296387},{"id":"https://openalex.org/keywords/theoretical-computer-science","display_name":"Theoretical computer science","score":0.3603709936141968},{"id":"https://openalex.org/keywords/computational-science","display_name":"Computational science","score":0.3471977114677429},{"id":"https://openalex.org/keywords/programming-language","display_name":"Programming language","score":0.19954782724380493},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.0783393383026123}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7608030438423157},{"id":"https://openalex.org/C43521106","wikidata":"https://www.wikidata.org/wiki/Q2165493","display_name":"Pipeline (software)","level":2,"score":0.7069889903068542},{"id":"https://openalex.org/C192209626","wikidata":"https://www.wikidata.org/wiki/Q190909","display_name":"Focus (optics)","level":2,"score":0.7049645185470581},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.630073606967926},{"id":"https://openalex.org/C168167062","wikidata":"https://www.wikidata.org/wiki/Q1117970","display_name":"Component (thermodynamics)","level":2,"score":0.6056195497512817},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.5670874118804932},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4663008749485016},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.45264485478401184},{"id":"https://openalex.org/C155281189","wikidata":"https://www.wikidata.org/wiki/Q3518150","display_name":"Tensor (intrinsic definition)","level":2,"score":0.4414559602737427},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.36550402641296387},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.3603709936141968},{"id":"https://openalex.org/C459310","wikidata":"https://www.wikidata.org/wiki/Q117801","display_name":"Computational science","level":1,"score":0.3471977114677429},{"id":"https://openalex.org/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.19954782724380493},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.0783393383026123},{"id":"https://openalex.org/C120665830","wikidata":"https://www.wikidata.org/wiki/Q14620","display_name":"Optics","level":1,"score":0.0},{"id":"https://openalex.org/C97355855","wikidata":"https://www.wikidata.org/wiki/Q11473","display_name":"Thermodynamics","level":1,"score":0.0},{"id":"https://openalex.org/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","level":0,"score":0.0},{"id":"https://openalex.org/C202444582","wikidata":"https://www.wikidata.org/wiki/Q837863","display_name":"Pure mathematics","level":1,"score":0.0}],"mesh":[],"locations_count":3,"locations":[{"id":"doi:10.1145/3815424","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3815424","pdf_url":null,"source":{"id":"https://openalex.org/S26056741","display_name":"ACM Transactions on Architecture and Code Optimization","issn_l":"1544-3566","issn":["1544-3566","1544-3973"],"is_oa":true,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319798","host_organization_name":"Association for Computing Machinery","host_organization_lineage":["https://openalex.org/P4310319798"],"host_organization_lineage_names":["Association for Computing Machinery"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ACM Transactions on Architecture and Code Optimization","raw_type":"journal-article"},{"id":"pmh:oai:arXiv.org:2304.07613","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2304.07613","pdf_url":"https://arxiv.org/pdf/2304.07613","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"","raw_type":"text"},{"id":"doi:10.48550/arxiv.2304.07613","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2304.07613","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"doi:10.1145/3815424","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3815424","pdf_url":null,"source":{"id":"https://openalex.org/S26056741","display_name":"ACM Transactions on Architecture and Code Optimization","issn_l":"1544-3566","issn":["1544-3566","1544-3973"],"is_oa":true,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319798","host_organization_name":"Association for Computing Machinery","host_organization_lineage":["https://openalex.org/P4310319798"],"host_organization_lineage_names":["Association for Computing Machinery"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ACM Transactions on Architecture and Code Optimization","raw_type":"journal-article"},"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G5839583985","display_name":null,"funder_award_id":"955513 (MAELSTROM) and 101034126 (EU-Pilot)","funder_id":"https://openalex.org/F1100377486","funder_display_name":"European High Performance Computing Joint Undertaking"},{"id":"https://openalex.org/G6531846704","display_name":"Empowering Computational Science using Data-Centric Programming","funder_award_id":"185778","funder_id":"https://openalex.org/F4320320924","funder_display_name":"Schweizerischer Nationalfonds zur F\u00f6rderung der Wissenschaftlichen Forschung"}],"funders":[{"id":"https://openalex.org/F1100377486","display_name":"European High Performance Computing Joint Undertaking","ror":null},{"id":"https://openalex.org/F4320306076","display_name":"National Science Foundation","ror":"https://ror.org/021nxhr62"},{"id":"https://openalex.org/F4320320924","display_name":"Schweizerischer Nationalfonds zur F\u00f6rderung der Wissenschaftlichen Forschung","ror":"https://ror.org/00yjd3n13"},{"id":"https://openalex.org/F4320331518","display_name":"Centro Svizzero di Calcolo Scientifico","ror":null},{"id":"https://openalex.org/F4320331617","display_name":"National Supercomputing Centre Singapore","ror":null}],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W4375867731","https://openalex.org/W3037187668","https://openalex.org/W2055243143","https://openalex.org/W2611989081","https://openalex.org/W4234772502","https://openalex.org/W4230611425","https://openalex.org/W2731899572","https://openalex.org/W2380685755","https://openalex.org/W4294635752","https://openalex.org/W4304166257"],"abstract_inverted_index":{"As":[0],"deep":[1,13],"learning":[2],"models":[3],"grow,":[4],"sparsity":[5,31,89,104,187,213],"is":[6,78,100],"becoming":[7],"an":[8],"increasingly":[9],"critical":[10],"component":[11],"of":[12,103,159,174,210],"neural":[14],"networks,":[15],"enabling":[16,146],"improved":[17],"performance":[18,151,207],"and":[19,49,80,92,107,118,130,208],"reduced":[20],"storage.":[21],"However,":[22],"existing":[23,64,163,217],"frameworks":[24,40,65],"offer":[25],"poor":[26],"support":[27,140],"for":[28,56,94,152,189,216],"sparsity.":[29],"Specialized":[30],"engines":[32],"focus":[33,42],"exclusively":[34],"on":[35,43],"sparse":[36,44,58,73,128],"inference,":[37],"while":[38,145],"general":[39],"primarily":[41],"tensors":[45],"in":[46],"classical":[47],"formats":[48],"neglect":[50],"the":[51,101,125,157],"broader":[52],"sparsification":[53,143,164],"pipeline":[54],"necessary":[55],"using":[57],"models,":[59],"especially":[60],"during":[61],"training.":[62],"Further,":[63],"are":[66],"not":[67],"easily":[68],"extensible:":[69],"adding":[70],"a":[71,88,170,180],"new":[72],"tensor":[74],"format":[75],"or":[76],"operator":[77],"challenging":[79],"time-consuming.":[81],"To":[82],"address":[83],"this,":[84],"we":[85,178],"propose":[86],"STen,":[87],"programming":[90],"model":[91],"interface":[93],"PyTorch":[95,218],"whose":[96],"key":[97],"design":[98],"insight":[99],"decoupling":[102],"layouts,":[105],"operators,":[106],"sparsifiers":[108],"into":[109],"composable,":[110],"first-class":[111],"abstractions":[112],"that":[113],"users":[114],"can":[115],"independently":[116],"define":[117],"combine.":[119],"An":[120],"automatic":[121],"dispatch":[122],"mechanism":[123],"selects":[124],"best":[126],"available":[127],"implementation":[129],"transparently":[131],"falls":[132],"back":[133],"to":[134,139,202],"dense":[135],"execution,":[136],"allowing":[137],"STen":[138,160,204],"virtually":[141],"all":[142],"methods":[144],"rapid":[147],"prototyping":[148],"without":[149],"sacrificing":[150],"optimized":[153],"paths.":[154],"We":[155],"demonstrate":[156],"versatility":[158],"by":[161,200],"expressing":[162],"techniques":[165],"within":[166],"its":[167],"abstraction,":[168],"achieving":[169],"code":[171],"size":[172],"reduction":[173],"over":[175],"2\u00d7.":[176],"Finally,":[177],"develop":[179],"novel,":[181],"high-performance":[182],"grouped":[183],"n":[184],":":[185],"m":[186],"layout":[188],"CPU":[190],"inference":[191,199],"at":[192],"moderate":[193],"sparsity,":[194],"accelerating":[195],"end-to-end":[196],"BERT":[197],"BASE":[198],"up":[201],"3.2\u00d7.":[203],"brings":[205],"high":[206],"ease":[209],"use,":[211],"making":[212],"readily":[214],"accessible":[215],"models.":[219]},"counts_by_year":[{"year":2026,"cited_by_count":1},{"year":2024,"cited_by_count":1}],"updated_date":"2026-07-21T08:15:58.654021","created_date":"2025-10-10T00:00:00"}