{"id":"https://openalex.org/W4388093538","doi":"https://doi.org/10.48550/arxiv.2310.19509","title":"SparseByteNN: A Novel Mobile Inference Acceleration Framework Based on Fine-Grained Group Sparsity","display_name":"SparseByteNN: A Novel Mobile Inference Acceleration Framework Based on Fine-Grained Group Sparsity","publication_year":2023,"publication_date":"2023-10-30","ids":{"openalex":"https://openalex.org/W4388093538","doi":"https://doi.org/10.48550/arxiv.2310.19509"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2310.19509","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2310.19509","pdf_url":"https://arxiv.org/pdf/2310.19509","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"","raw_type":"text"},"type":"preprint","indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2310.19509","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5100928937","display_name":"Haitao Xu","orcid":null},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Xu, Haitao","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5041983865","display_name":"Songwei Liu","orcid":"https://orcid.org/0000-0002-9892-0918"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Liu, Songwei","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5074623897","display_name":"Yuyang Xu","orcid":"https://orcid.org/0000-0002-3962-3526"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xu, Yuyang","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100328245","display_name":"Shuai Wang","orcid":"https://orcid.org/0000-0001-8897-9476"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Shuai","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5014851555","display_name":"Jiashi Li","orcid":"https://orcid.org/0009-0004-7171-8527"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Li, Jiashi","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5048876463","display_name":"Chenqian Yan","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yan, Chenqian","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":null,"display_name":"Li, Liangqiang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Li, Liangqiang","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5073867140","display_name":"Lean Fu","orcid":"https://orcid.org/0000-0003-1889-9352"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Fu, Lean","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5089480544","display_name":"Xin Pan","orcid":"https://orcid.org/0000-0003-1437-2780"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Pan, Xin","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5078392945","display_name":"Fangmin Chen","orcid":"https://orcid.org/0009-0002-9613-7898"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chen, Fangmin","raw_affiliation_strings":[],"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":10,"corresponding_author_ids":["https://openalex.org/A5100928937"],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"cited_by_count":0,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":null,"last_page":null},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.9980000257492065,"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":0.9980000257492065,"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/T12676","display_name":"Machine Learning and ELM","score":0.9916999936103821,"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/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.9854999780654907,"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/speedup","display_name":"Speedup","score":0.8498666882514954},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.8211961984634399},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.7829047441482544},{"id":"https://openalex.org/keywords/pruning","display_name":"Pruning","score":0.6338698863983154},{"id":"https://openalex.org/keywords/granularity","display_name":"Granularity","score":0.5814985036849976},{"id":"https://openalex.org/keywords/kernel","display_name":"Kernel (algebra)","score":0.5526663064956665},{"id":"https://openalex.org/keywords/flops","display_name":"FLOPS","score":0.4945504367351532},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.4711902141571045},{"id":"https://openalex.org/keywords/sparse-matrix","display_name":"Sparse matrix","score":0.4678027927875519},{"id":"https://openalex.org/keywords/schema","display_name":"Schema (genetic algorithms)","score":0.415947288274765},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.397617906332016},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.27459901571273804},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.22784408926963806}],"concepts":[{"id":"https://openalex.org/C68339613","wikidata":"https://www.wikidata.org/wiki/Q1549489","display_name":"Speedup","level":2,"score":0.8498666882514954},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8211961984634399},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.7829047441482544},{"id":"https://openalex.org/C108010975","wikidata":"https://www.wikidata.org/wiki/Q500094","display_name":"Pruning","level":2,"score":0.6338698863983154},{"id":"https://openalex.org/C177774035","wikidata":"https://www.wikidata.org/wiki/Q1246948","display_name":"Granularity","level":2,"score":0.5814985036849976},{"id":"https://openalex.org/C74193536","wikidata":"https://www.wikidata.org/wiki/Q574844","display_name":"Kernel (algebra)","level":2,"score":0.5526663064956665},{"id":"https://openalex.org/C3826847","wikidata":"https://www.wikidata.org/wiki/Q188768","display_name":"FLOPS","level":2,"score":0.4945504367351532},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.4711902141571045},{"id":"https://openalex.org/C56372850","wikidata":"https://www.wikidata.org/wiki/Q1050404","display_name":"Sparse matrix","level":3,"score":0.4678027927875519},{"id":"https://openalex.org/C52146309","wikidata":"https://www.wikidata.org/wiki/Q7431116","display_name":"Schema (genetic algorithms)","level":2,"score":0.415947288274765},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.397617906332016},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.27459901571273804},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.22784408926963806},{"id":"https://openalex.org/C86803240","wikidata":"https://www.wikidata.org/wiki/Q420","display_name":"Biology","level":0,"score":0.0},{"id":"https://openalex.org/C6557445","wikidata":"https://www.wikidata.org/wiki/Q173113","display_name":"Agronomy","level":1,"score":0.0},{"id":"https://openalex.org/C163716315","wikidata":"https://www.wikidata.org/wiki/Q901177","display_name":"Gaussian","level":2,"score":0.0},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.0},{"id":"https://openalex.org/C114614502","wikidata":"https://www.wikidata.org/wiki/Q76592","display_name":"Combinatorics","level":1,"score":0.0},{"id":"https://openalex.org/C62520636","wikidata":"https://www.wikidata.org/wiki/Q944","display_name":"Quantum mechanics","level":1,"score":0.0},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","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}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2310.19509","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2310.19509","pdf_url":"https://arxiv.org/pdf/2310.19509","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"","raw_type":"text"},{"id":"doi:10.48550/arxiv.2310.19509","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2310.19509","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"article"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2310.19509","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2310.19509","pdf_url":"https://arxiv.org/pdf/2310.19509","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"","raw_type":"text"},"sustainable_development_goals":[{"display_name":"Industry, innovation and infrastructure","score":0.44999998807907104,"id":"https://metadata.un.org/sdg/9"}],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4388093538.pdf","grobid_xml":"https://content.openalex.org/works/W4388093538.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W2058965144","https://openalex.org/W2164382479","https://openalex.org/W2146343568","https://openalex.org/W98480971","https://openalex.org/W2150291671","https://openalex.org/W2013643406","https://openalex.org/W2027972911","https://openalex.org/W2157978810","https://openalex.org/W1969153118","https://openalex.org/W2891818448"],"abstract_inverted_index":{"To":[0],"address":[1],"the":[2,96,107,115,173,192,199],"challenge":[3],"of":[4,61,145,158,166,210,215],"increasing":[5],"network":[6,14,93],"size,":[7],"researchers":[8],"have":[9],"developed":[10],"sparse":[11,83,116,147,159,164,185,201],"models":[12],"through":[13],"pruning.":[15,79],"However,":[16],"maintaining":[17],"model":[18],"accuracy":[19,208],"while":[20],"achieving":[21],"significant":[22],"speedups":[23],"on":[24,172,178],"general":[25],"computing":[26],"devices":[27],"remains":[28],"an":[29],"open":[30],"problem.":[31],"In":[32],"this":[33,123,139],"paper,":[34],"we":[35,161],"present":[36],"a":[37,71,99,143,206],"novel":[38],"mobile":[39],"inference":[40,202],"acceleration":[41],"framework":[42,59],"SparseByteNN,":[43],"which":[44],"leverages":[45],"fine-grained":[46,66],"kernel":[47,67],"sparsity":[48,68,72],"to":[49,137],"achieve":[50],"real-time":[51],"execution":[52],"as":[53,55],"well":[54],"high":[56,100,104],"accuracy.":[57],"Our":[58],"consists":[60],"two":[62],"parts:":[63],"(a)":[64],"A":[65],"schema":[69,97],"with":[70,89,114,154,205],"granularity":[73],"between":[74],"structured":[75],"pruning":[76],"and":[77,103,151,195],"unstructured":[78],"It":[80],"designs":[81],"multiple":[82],"patterns":[84],"for":[85,149,183],"different":[86],"operators.":[87],"Combined":[88],"our":[90,155],"proposed":[91],"whole":[92],"rearrangement":[94],"strategy,":[95],"achieves":[98,188],"compression":[101],"rate":[102],"precision":[105],"at":[106,220],"same":[108],"time.":[109],"(b)":[110],"Inference":[111],"engine":[112,203],"co-optimized":[113],"pattern.":[117],"The":[118,212],"conventional":[119],"wisdom":[120],"is":[121],"that":[122,163,182],"reduction":[124],"in":[125],"theoretical":[126],"FLOPs":[127],"does":[128],"not":[129],"translate":[130],"into":[131],"real-world":[132],"efficiency":[133],"gains.":[134],"We":[135],"aim":[136],"correct":[138],"misconception":[140],"by":[141],"introducing":[142],"family":[144],"efficient":[146,156],"kernels":[148],"ARM":[150],"WebAssembly.":[152],"Equipped":[153],"implementation":[157],"primitives,":[160],"show":[162,181],"versions":[165],"MobileNet-v1":[167],"outperform":[168],"strong":[169],"dense":[170,193],"baselines":[171],"efficiency-accuracy":[174],"curve.":[175],"Experimental":[176],"results":[177],"Qualcomm":[179],"855":[180],"30%":[184],"MobileNet-v1,":[186],"SparseByteNN":[187,216],"1.27x":[189],"speedup":[190,197],"over":[191,198],"version":[194],"1.29x":[196],"state-of-the-art":[200],"MNN":[204],"slight":[207],"drop":[209],"0.224%.":[211],"source":[213],"code":[214],"will":[217],"be":[218],"available":[219],"https://github.com/lswzjuer/SparseByteNN":[221]},"counts_by_year":[],"updated_date":"2026-03-08T08:50:53.379069","created_date":"2023-11-01T00:00:00"}