{"id":"https://openalex.org/W4409248508","doi":"https://doi.org/10.1109/hpca61900.2025.00076","title":"CROSS: Compiler-Driven Optimization of Sparse DNNs Using Sparse/Dense Computation Kernels","display_name":"CROSS: Compiler-Driven Optimization of Sparse DNNs Using Sparse/Dense Computation Kernels","publication_year":2025,"publication_date":"2025-03-01","ids":{"openalex":"https://openalex.org/W4409248508","doi":"https://doi.org/10.1109/hpca61900.2025.00076"},"language":"en","primary_location":{"id":"doi:10.1109/hpca61900.2025.00076","is_oa":false,"landing_page_url":"https://doi.org/10.1109/hpca61900.2025.00076","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 IEEE International Symposium on High Performance Computer Architecture (HPCA)","raw_type":"proceedings-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":"https://openalex.org/A5017670541","display_name":"Fangxin Liu","orcid":"https://orcid.org/0000-0002-8769-293X"},"institutions":[{"id":"https://openalex.org/I183067930","display_name":"Shanghai Jiao Tong University","ror":"https://ror.org/0220qvk04","country_code":"CN","type":"education","lineage":["https://openalex.org/I183067930"]}],"countries":["CN"],"is_corresponding":true,"raw_author_name":"Fangxin Liu","raw_affiliation_strings":["Shanghai Jiao Tong University,Department of Computer Science and Engineering"],"affiliations":[{"raw_affiliation_string":"Shanghai Jiao Tong University,Department of Computer Science and Engineering","institution_ids":["https://openalex.org/I183067930"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5102989014","display_name":"Shiyuan Huang","orcid":"https://orcid.org/0000-0002-8086-6802"},"institutions":[{"id":"https://openalex.org/I183067930","display_name":"Shanghai Jiao Tong University","ror":"https://ror.org/0220qvk04","country_code":"CN","type":"education","lineage":["https://openalex.org/I183067930"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Shiyuan Huang","raw_affiliation_strings":["Shanghai Jiao Tong University,Department of Computer Science and Engineering"],"affiliations":[{"raw_affiliation_string":"Shanghai Jiao Tong University,Department of Computer Science and Engineering","institution_ids":["https://openalex.org/I183067930"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5030493839","display_name":"Ning Yang","orcid":"https://orcid.org/0009-0004-6964-8910"},"institutions":[{"id":"https://openalex.org/I183067930","display_name":"Shanghai Jiao Tong University","ror":"https://ror.org/0220qvk04","country_code":"CN","type":"education","lineage":["https://openalex.org/I183067930"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Ning Yang","raw_affiliation_strings":["Shanghai Jiao Tong University,Department of Computer Science and Engineering"],"affiliations":[{"raw_affiliation_string":"Shanghai Jiao Tong University,Department of Computer Science and Engineering","institution_ids":["https://openalex.org/I183067930"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5107269495","display_name":"Zongwu Wang","orcid":"https://orcid.org/0009-0003-2157-4927"},"institutions":[{"id":"https://openalex.org/I183067930","display_name":"Shanghai Jiao Tong University","ror":"https://ror.org/0220qvk04","country_code":"CN","type":"education","lineage":["https://openalex.org/I183067930"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zongwu Wang","raw_affiliation_strings":["Shanghai Jiao Tong University,Department of Computer Science and Engineering"],"affiliations":[{"raw_affiliation_string":"Shanghai Jiao Tong University,Department of Computer Science and Engineering","institution_ids":["https://openalex.org/I183067930"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101399528","display_name":"Haoming Li","orcid":"https://orcid.org/0000-0002-2939-6534"},"institutions":[{"id":"https://openalex.org/I183067930","display_name":"Shanghai Jiao Tong University","ror":"https://ror.org/0220qvk04","country_code":"CN","type":"education","lineage":["https://openalex.org/I183067930"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Haomin Li","raw_affiliation_strings":["Shanghai Jiao Tong University,Department of Computer Science and Engineering"],"affiliations":[{"raw_affiliation_string":"Shanghai Jiao Tong University,Department of Computer Science and Engineering","institution_ids":["https://openalex.org/I183067930"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5053801300","display_name":"Li Jiang","orcid":"https://orcid.org/0000-0002-7353-8798"},"institutions":[{"id":"https://openalex.org/I183067930","display_name":"Shanghai Jiao Tong University","ror":"https://ror.org/0220qvk04","country_code":"CN","type":"education","lineage":["https://openalex.org/I183067930"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Li Jiang","raw_affiliation_strings":["Shanghai Jiao Tong University,Department of Computer Science and Engineering"],"affiliations":[{"raw_affiliation_string":"Shanghai Jiao Tong University,Department of Computer Science and Engineering","institution_ids":["https://openalex.org/I183067930"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":6,"corresponding_author_ids":["https://openalex.org/A5017670541"],"corresponding_institution_ids":["https://openalex.org/I183067930"],"apc_list":null,"apc_paid":null,"fwci":1.8193,"has_fulltext":false,"cited_by_count":2,"citation_normalized_percentile":{"value":0.8424459,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":95,"max":96},"biblio":{"volume":null,"issue":null,"first_page":"963","last_page":"976"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T12702","display_name":"Brain Tumor Detection and Classification","score":0.9941999912261963,"subfield":{"id":"https://openalex.org/subfields/2808","display_name":"Neurology"},"field":{"id":"https://openalex.org/fields/28","display_name":"Neuroscience"},"domain":{"id":"https://openalex.org/domains/1","display_name":"Life Sciences"}},"topics":[{"id":"https://openalex.org/T12702","display_name":"Brain Tumor Detection and Classification","score":0.9941999912261963,"subfield":{"id":"https://openalex.org/subfields/2808","display_name":"Neurology"},"field":{"id":"https://openalex.org/fields/28","display_name":"Neuroscience"},"domain":{"id":"https://openalex.org/domains/1","display_name":"Life Sciences"}},{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.9937999844551086,"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/T10812","display_name":"Human Pose and Action Recognition","score":0.9524000287055969,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.7637431621551514},{"id":"https://openalex.org/keywords/compiler","display_name":"Compiler","score":0.7510796785354614},{"id":"https://openalex.org/keywords/computation","display_name":"Computation","score":0.6220554113388062},{"id":"https://openalex.org/keywords/sparse-array","display_name":"Sparse array","score":0.6003354787826538},{"id":"https://openalex.org/keywords/sparse-matrix","display_name":"Sparse matrix","score":0.5961484909057617},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.5678893327713013},{"id":"https://openalex.org/keywords/optimizing-compiler","display_name":"Optimizing compiler","score":0.47884368896484375},{"id":"https://openalex.org/keywords/sparse-approximation","display_name":"Sparse approximation","score":0.4281757175922394},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.2907906174659729},{"id":"https://openalex.org/keywords/programming-language","display_name":"Programming language","score":0.16068023443222046},{"id":"https://openalex.org/keywords/physics","display_name":"Physics","score":0.09372377395629883}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7637431621551514},{"id":"https://openalex.org/C169590947","wikidata":"https://www.wikidata.org/wiki/Q47506","display_name":"Compiler","level":2,"score":0.7510796785354614},{"id":"https://openalex.org/C45374587","wikidata":"https://www.wikidata.org/wiki/Q12525525","display_name":"Computation","level":2,"score":0.6220554113388062},{"id":"https://openalex.org/C145177509","wikidata":"https://www.wikidata.org/wiki/Q1050404","display_name":"Sparse array","level":2,"score":0.6003354787826538},{"id":"https://openalex.org/C56372850","wikidata":"https://www.wikidata.org/wiki/Q1050404","display_name":"Sparse matrix","level":3,"score":0.5961484909057617},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.5678893327713013},{"id":"https://openalex.org/C190902152","wikidata":"https://www.wikidata.org/wiki/Q1325106","display_name":"Optimizing compiler","level":3,"score":0.47884368896484375},{"id":"https://openalex.org/C124066611","wikidata":"https://www.wikidata.org/wiki/Q28684319","display_name":"Sparse approximation","level":2,"score":0.4281757175922394},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.2907906174659729},{"id":"https://openalex.org/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.16068023443222046},{"id":"https://openalex.org/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","level":0,"score":0.09372377395629883},{"id":"https://openalex.org/C163716315","wikidata":"https://www.wikidata.org/wiki/Q901177","display_name":"Gaussian","level":2,"score":0.0},{"id":"https://openalex.org/C62520636","wikidata":"https://www.wikidata.org/wiki/Q944","display_name":"Quantum mechanics","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/hpca61900.2025.00076","is_oa":false,"landing_page_url":"https://doi.org/10.1109/hpca61900.2025.00076","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 IEEE International Symposium on High Performance Computer Architecture (HPCA)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[{"id":"https://openalex.org/F4320321001","display_name":"National Natural Science Foundation of China","ror":"https://ror.org/01h0zpd94"}],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":53,"referenced_works":["https://openalex.org/W1519855838","https://openalex.org/W1573728792","https://openalex.org/W2009654791","https://openalex.org/W2066155701","https://openalex.org/W2099625934","https://openalex.org/W2117539524","https://openalex.org/W2168809123","https://openalex.org/W2194775991","https://openalex.org/W2278832452","https://openalex.org/W2590246587","https://openalex.org/W2898123186","https://openalex.org/W2914631005","https://openalex.org/W2923014074","https://openalex.org/W2951178714","https://openalex.org/W2964337156","https://openalex.org/W2980186997","https://openalex.org/W3012561096","https://openalex.org/W3095319910","https://openalex.org/W3130660608","https://openalex.org/W3132695675","https://openalex.org/W3133862166","https://openalex.org/W3158112431","https://openalex.org/W3159727696","https://openalex.org/W3204021316","https://openalex.org/W3206453033","https://openalex.org/W3212488865","https://openalex.org/W4255450819","https://openalex.org/W4285104934","https://openalex.org/W4293024985","https://openalex.org/W4312191446","https://openalex.org/W4327911434","https://openalex.org/W4383753658","https://openalex.org/W4393145428","https://openalex.org/W4393406920","https://openalex.org/W4403407616","https://openalex.org/W4404954642","https://openalex.org/W6637373629","https://openalex.org/W6725543821","https://openalex.org/W6738642365","https://openalex.org/W6755120062","https://openalex.org/W6755207826","https://openalex.org/W6769311223","https://openalex.org/W6779728309","https://openalex.org/W6784999070","https://openalex.org/W6787972765","https://openalex.org/W6796549874","https://openalex.org/W6797677657","https://openalex.org/W6803366298","https://openalex.org/W6803815625","https://openalex.org/W6850625674","https://openalex.org/W6854866820","https://openalex.org/W6857251794","https://openalex.org/W6862046771"],"related_works":["https://openalex.org/W2002505081","https://openalex.org/W2083681681","https://openalex.org/W2577630842","https://openalex.org/W1172579163","https://openalex.org/W1488300410","https://openalex.org/W2138790427","https://openalex.org/W3146558274","https://openalex.org/W4246454774","https://openalex.org/W2162051035","https://openalex.org/W2051410394"],"abstract_inverted_index":{"As":[0],"deep":[1],"learning":[2],"models":[3,189],"continue":[4],"to":[5,34,58,74,89,129,148,179],"grow":[6],"larger":[7],"and":[8,24,43,61,94,166,190],"more":[9,35],"complex,":[10],"exploiting":[11],"sparsity":[12,30,137,164],"is":[13,88],"becoming":[14],"one":[15],"of":[16,112,153],"the":[17,38,96,101,123,154],"most":[18],"critical":[19],"areas":[20],"for":[21,28],"enhancing":[22],"efficiency":[23],"scalability.":[25],"Several":[26],"methods":[27,47],"leveraging":[29],"have":[31],"been":[32],"proposed":[33],"effectively":[36],"balance":[37],"trade-off":[39],"between":[40],"compression":[41],"ratio":[42],"accuracy.":[44],"While":[45],"these":[46],"offer":[48],"algorithmic":[49],"advantages,":[50],"they":[51],"also":[52],"introduce":[53,141],"significant":[54],"hardware":[55],"overhead":[56],"due":[57],"index-based":[59],"encoding":[60],"decoding.":[62],"In":[63],"this":[64],"paper,":[65],"we":[66,107,140],"propose":[67],"CROSS,":[68],"an":[69,109,170],"end-to-end":[70],"compilation":[71],"optimization":[72,167],"technique":[73,147],"achieve":[75],"sparse":[76,113,124,144,155],"DNN":[77,98,117],"acceleration":[78],"using":[79],"GPU":[80],"computation":[81,99,104,150],"kernels.":[82,105],"The":[83,157],"key":[84],"insight":[85],"behind":[86],"CROSS":[87,159],"exploit":[90],"parameter":[91],"distribution":[92],"locality":[93],"reconcile":[95],"\u201csparse\u201d":[97],"with":[100,135,183],"high-performance":[102],"\u201cdense\u201d":[103],"Specifically,":[106],"perform":[108],"in-depth":[110],"analysis":[111],"operations":[114],"in":[115],"mainstream":[116],"computing":[118],"frameworks.":[119],"We":[120],"then":[121],"decompose":[122],"workload":[125],"into":[126],"multiple":[127],"components":[128],"create":[130],"highly":[131],"efficient,":[132],"specialized":[133],"operators":[134],"different":[136],"levels.":[138],"Additionally,":[139],"a":[142],"novel":[143],"graph":[145],"translation":[146],"facilitate":[149],"kernel":[151],"processing":[152],"workload.":[156],"resulting":[158],"framework":[160],"can":[161],"accommodate":[162],"various":[163,188],"patterns":[165],"techniques,":[168],"delivering":[169],"average":[171],"$2.03":[172],"\\times$":[173],"speedup":[174],"on":[175],"inference":[176],"latency":[177],"compared":[178],"seven":[180],"state-of-the-art":[181],"solutions":[182],"smaller":[184],"memory":[185],"footprints":[186],"across":[187],"datasets.":[191]},"counts_by_year":[{"year":2025,"cited_by_count":2}],"updated_date":"2025-11-06T03:46:38.306776","created_date":"2025-10-10T00:00:00"}