{"id":"https://openalex.org/W4413361526","doi":"https://doi.org/10.1109/asap65064.2025.00016","title":"Trimming Down Large Spiking Vision Transformers Via Heterogeneous Quantization Search","display_name":"Trimming Down Large Spiking Vision Transformers Via Heterogeneous Quantization Search","publication_year":2025,"publication_date":"2025-07-28","ids":{"openalex":"https://openalex.org/W4413361526","doi":"https://doi.org/10.1109/asap65064.2025.00016"},"language":"en","primary_location":{"id":"doi:10.1109/asap65064.2025.00016","is_oa":false,"landing_page_url":"https://doi.org/10.1109/asap65064.2025.00016","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 IEEE 36th International Conference on Application-specific Systems, Architectures and Processors (ASAP)","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/A5101268560","display_name":"Boxun Xu","orcid":"https://orcid.org/0009-0003-2896-6632"},"institutions":[{"id":"https://openalex.org/I126031752","display_name":"Santa Barbara City College","ror":"https://ror.org/022j4a796","country_code":"US","type":"education","lineage":["https://openalex.org/I126031752"]},{"id":"https://openalex.org/I154570441","display_name":"University of California, Santa Barbara","ror":"https://ror.org/02t274463","country_code":"US","type":"education","lineage":["https://openalex.org/I154570441"]}],"countries":["US"],"is_corresponding":true,"raw_author_name":"Boxun Xu","raw_affiliation_strings":["UC Santa Barbara,ECE Department,Santa Barbara,CA,USA"],"affiliations":[{"raw_affiliation_string":"UC Santa Barbara,ECE Department,Santa Barbara,CA,USA","institution_ids":["https://openalex.org/I154570441","https://openalex.org/I126031752"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5112618994","display_name":"Yufei Song","orcid":null},"institutions":[{"id":"https://openalex.org/I154570441","display_name":"University of California, Santa Barbara","ror":"https://ror.org/02t274463","country_code":"US","type":"education","lineage":["https://openalex.org/I154570441"]},{"id":"https://openalex.org/I126031752","display_name":"Santa Barbara City College","ror":"https://ror.org/022j4a796","country_code":"US","type":"education","lineage":["https://openalex.org/I126031752"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Yufei Song","raw_affiliation_strings":["UC Santa Barbara,ECE Department,Santa Barbara,CA,USA"],"affiliations":[{"raw_affiliation_string":"UC Santa Barbara,ECE Department,Santa Barbara,CA,USA","institution_ids":["https://openalex.org/I154570441","https://openalex.org/I126031752"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5100432782","display_name":"Peng Li","orcid":"https://orcid.org/0000-0003-3548-4589"},"institutions":[{"id":"https://openalex.org/I126031752","display_name":"Santa Barbara City College","ror":"https://ror.org/022j4a796","country_code":"US","type":"education","lineage":["https://openalex.org/I126031752"]},{"id":"https://openalex.org/I154570441","display_name":"University of California, Santa Barbara","ror":"https://ror.org/02t274463","country_code":"US","type":"education","lineage":["https://openalex.org/I154570441"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Peng Li","raw_affiliation_strings":["UC Santa Barbara,ECE Department,Santa Barbara,CA,USA"],"affiliations":[{"raw_affiliation_string":"UC Santa Barbara,ECE Department,Santa Barbara,CA,USA","institution_ids":["https://openalex.org/I154570441","https://openalex.org/I126031752"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":3,"corresponding_author_ids":["https://openalex.org/A5101268560"],"corresponding_institution_ids":["https://openalex.org/I126031752","https://openalex.org/I154570441"],"apc_list":null,"apc_paid":null,"fwci":0.7491,"has_fulltext":false,"cited_by_count":1,"citation_normalized_percentile":{"value":0.76755141,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":91,"max":95},"biblio":{"volume":null,"issue":null,"first_page":"41","last_page":"48"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10502","display_name":"Advanced Memory and Neural Computing","score":0.9973999857902527,"subfield":{"id":"https://openalex.org/subfields/2208","display_name":"Electrical and Electronic Engineering"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T10502","display_name":"Advanced Memory and Neural Computing","score":0.9973999857902527,"subfield":{"id":"https://openalex.org/subfields/2208","display_name":"Electrical and Electronic Engineering"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T11992","display_name":"CCD and CMOS Imaging Sensors","score":0.9907000064849854,"subfield":{"id":"https://openalex.org/subfields/2208","display_name":"Electrical and Electronic Engineering"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10320","display_name":"Neural Networks and Applications","score":0.9544000029563904,"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/trimming","display_name":"Trimming","score":0.9610632658004761},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.5600141286849976},{"id":"https://openalex.org/keywords/transformer","display_name":"Transformer","score":0.5054197311401367},{"id":"https://openalex.org/keywords/quantization","display_name":"Quantization (signal processing)","score":0.4912840723991394},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.46183159947395325},{"id":"https://openalex.org/keywords/computer-vision","display_name":"Computer vision","score":0.41856369376182556},{"id":"https://openalex.org/keywords/engineering","display_name":"Engineering","score":0.18027305603027344},{"id":"https://openalex.org/keywords/electrical-engineering","display_name":"Electrical engineering","score":0.13261067867279053},{"id":"https://openalex.org/keywords/voltage","display_name":"Voltage","score":0.09174394607543945}],"concepts":[{"id":"https://openalex.org/C56951928","wikidata":"https://www.wikidata.org/wiki/Q3539213","display_name":"Trimming","level":2,"score":0.9610632658004761},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5600141286849976},{"id":"https://openalex.org/C66322947","wikidata":"https://www.wikidata.org/wiki/Q11658","display_name":"Transformer","level":3,"score":0.5054197311401367},{"id":"https://openalex.org/C28855332","wikidata":"https://www.wikidata.org/wiki/Q198099","display_name":"Quantization (signal processing)","level":2,"score":0.4912840723991394},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.46183159947395325},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.41856369376182556},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.18027305603027344},{"id":"https://openalex.org/C119599485","wikidata":"https://www.wikidata.org/wiki/Q43035","display_name":"Electrical engineering","level":1,"score":0.13261067867279053},{"id":"https://openalex.org/C165801399","wikidata":"https://www.wikidata.org/wiki/Q25428","display_name":"Voltage","level":2,"score":0.09174394607543945},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/asap65064.2025.00016","is_oa":false,"landing_page_url":"https://doi.org/10.1109/asap65064.2025.00016","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 IEEE 36th International Conference on Application-specific Systems, Architectures and Processors (ASAP)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[{"score":0.41999998688697815,"display_name":"Affordable and clean energy","id":"https://metadata.un.org/sdg/7"}],"awards":[{"id":"https://openalex.org/G2283441041","display_name":null,"funder_award_id":"1948201","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"}],"funders":[{"id":"https://openalex.org/F4320306076","display_name":"National Science Foundation","ror":"https://ror.org/021nxhr62"}],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":13,"referenced_works":["https://openalex.org/W1992348535","https://openalex.org/W2004528524","https://openalex.org/W2120714355","https://openalex.org/W2138913040","https://openalex.org/W2619510810","https://openalex.org/W2783525259","https://openalex.org/W2944996566","https://openalex.org/W2962804204","https://openalex.org/W3199910846","https://openalex.org/W4280536985","https://openalex.org/W4379115853","https://openalex.org/W4409285493","https://openalex.org/W4411486515"],"related_works":["https://openalex.org/W2772917594","https://openalex.org/W2036807459","https://openalex.org/W2058170566","https://openalex.org/W2755342338","https://openalex.org/W2166024367","https://openalex.org/W3116076068","https://openalex.org/W2229312674","https://openalex.org/W2951359407","https://openalex.org/W2079911747","https://openalex.org/W1969923398"],"abstract_inverted_index":{"Spiking":[0],"neural":[1,35],"networks":[2,36],"(SNNs)":[3],"are":[4],"amenable":[5],"to":[6,15,30,81],"deployment":[7],"on":[8,48,144,186],"edge":[9,50],"devices":[10,51,55],"and":[11,148,181,190,206],"neuromorphic":[12,145],"hardware":[13],"due":[14],"their":[16,31],"lower":[17],"dissipation.":[18],"Recently,":[19],"SNN-based":[20],"transformers":[21,84,124],"have":[22],"garnered":[23],"significant":[24,58],"interest,":[25],"incorporating":[26],"attention":[27],"mechanisms":[28],"akin":[29],"counterparts":[32],"in":[33],"artificial":[34],"(ANNs)":[37],"while":[38,125,194],"demonstrating":[39],"excellent":[40],"performance.":[41],"However,":[42],"deploying":[43],"large-scale":[44],"spiking":[45,83,123,164],"transformer":[46],"models":[47],"resourceconstrained":[49],"such":[52],"as":[53],"mobile":[54],"still":[56],"poses":[57],"challenges":[59],"resulting":[60],"from":[61],"the":[62,91,116,167,187],"high":[63,120,196],"computational":[64],"demands":[65],"of":[66,93,98,118,131,174,199],"large":[67],"uncompressed":[68],"high-precision":[69],"models.":[70],"In":[71],"this":[72],"work,":[73],"we":[74],"introduce":[75],"a":[76,137],"novel":[77],"heterogeneous":[78,113],"quantization":[79,92,101,114],"method":[80],"compress":[82],"through":[85],"layer-wise":[86],"quantization.":[87],"Our":[88,112],"approach":[89,169],"optimizes":[90],"each":[94],"layer":[95],"using":[96],"one":[97],"two":[99],"distinct":[100],"schemes,":[102],"that":[103],"is,":[104],"uniform":[105],"or":[106],"power-of-two":[107],"quantization,":[108],"with":[109,134],"mixed":[110],"precision.":[111],"demonstrates":[115],"feasibility":[117],"maintaining":[119,195],"performance":[121],"for":[122,161],"utilizing":[126],"an":[127],"average":[128],"effective":[129],"precision":[130],"3.14-3.67":[132],"bits":[133],"less":[135],"than":[136],"<tex":[138,153,175,182,200],"xmlns:mml=\"http://www.w3.org/1998/Math/MathML\"":[139,154,176,183,201],"xmlns:xlink=\"http://www.w3.org/1999/xlink\">$\\mathbf{1":[140],"\\%}$</tex>":[141],"accuracy":[142,197],"drop":[143],"DVS":[146],"Gesture":[147],"CIFAR10-DVS":[149],"datasets.":[150],"It":[151],"achieves":[152,170],"xmlns:xlink=\"http://www.w3.org/1999/xlink\">$8.71":[155],"\\times-10.19":[156],"\\times$</tex>":[157,185],"model":[158],"compression":[159],"rate":[160],"standard":[162],"floating-point":[163],"transformers.":[165],"Furthermore,":[166],"proposed":[168],"substantial":[171],"energy":[172],"reductions":[173],"xmlns:xlink=\"http://www.w3.org/1999/xlink\">$5.69":[177],"\\times,":[178],"8.72":[179],"\\times$</tex>,":[180],"xmlns:xlink=\"http://www.w3.org/1999/xlink\">$10.2":[184],"N-Caltech101,":[188],"DVS-Gesture,":[189],"CIFAR10DVS":[191],"datasets,":[192],"respectively,":[193],"levels":[198],"xmlns:xlink=\"http://www.w3.org/1999/xlink\">$85.3":[202],"\\%,":[203],"97.57":[204],"\\%$</tex>,":[205],"80.4":[207],"%.":[208]},"counts_by_year":[{"year":2025,"cited_by_count":1}],"updated_date":"2025-11-06T03:46:38.306776","created_date":"2025-10-10T00:00:00"}