{"id":"https://openalex.org/W4297095249","doi":"https://doi.org/10.1109/tpami.2022.3208922","title":"Token Selection is a Simple Booster for Vision Transformers","display_name":"Token Selection is a Simple Booster for Vision Transformers","publication_year":2022,"publication_date":"2022-01-01","ids":{"openalex":"https://openalex.org/W4297095249","doi":"https://doi.org/10.1109/tpami.2022.3208922","pmid":"https://pubmed.ncbi.nlm.nih.gov/36155475"},"language":"en","primary_location":{"id":"doi:10.1109/tpami.2022.3208922","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tpami.2022.3208922","pdf_url":null,"source":{"id":"https://openalex.org/S199944782","display_name":"IEEE Transactions on Pattern Analysis and Machine Intelligence","issn_l":"0162-8828","issn":["0162-8828","1939-3539","2160-9292"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310320439","host_organization_name":"IEEE Computer Society","host_organization_lineage":["https://openalex.org/P4310320439","https://openalex.org/P4310319808"],"host_organization_lineage_names":["IEEE Computer Society","Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Transactions on Pattern Analysis and Machine Intelligence","raw_type":"journal-article"},"type":"article","indexed_in":["crossref","pubmed"],"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/A5072984967","display_name":"Daquan Zhou","orcid":"https://orcid.org/0000-0002-4771-1796"},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Daquan Zhou","raw_affiliation_strings":["Bytedance"],"affiliations":[{"raw_affiliation_string":"Bytedance","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5040392623","display_name":"Qibin Hou","orcid":"https://orcid.org/0000-0002-8388-8708"},"institutions":[{"id":"https://openalex.org/I205237279","display_name":"Nankai University","ror":"https://ror.org/01y1kjr75","country_code":"CN","type":"education","lineage":["https://openalex.org/I205237279"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Qibin Hou","raw_affiliation_strings":["School of Computer Science, Nankai University, Tianjin, China"],"affiliations":[{"raw_affiliation_string":"School of Computer Science, Nankai University, Tianjin, China","institution_ids":["https://openalex.org/I205237279"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5042544148","display_name":"Linjie Yang","orcid":"https://orcid.org/0000-0003-2766-1143"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Linjie Yang","raw_affiliation_strings":["ByteDance, USA"],"affiliations":[{"raw_affiliation_string":"ByteDance, USA","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100668699","display_name":"Xiaojie Jin","orcid":"https://orcid.org/0000-0002-7850-1353"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xiaojie Jin","raw_affiliation_strings":["ByteDance, USA"],"affiliations":[{"raw_affiliation_string":"ByteDance, USA","institution_ids":[]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5100668696","display_name":"Jiashi Feng","orcid":"https://orcid.org/0000-0001-6843-0064"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jiashi Feng","raw_affiliation_strings":["ByteDance, Singapore"],"affiliations":[{"raw_affiliation_string":"ByteDance, Singapore","institution_ids":[]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":5,"corresponding_author_ids":["https://openalex.org/A5072984967"],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":1.3086,"has_fulltext":false,"cited_by_count":13,"citation_normalized_percentile":{"value":0.81548296,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":94,"max":99},"biblio":{"volume":"45","issue":"11","first_page":"1","last_page":"9"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.9998999834060669,"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":0.9998999834060669,"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/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.9993000030517578,"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/T10627","display_name":"Advanced Image and Video Retrieval Techniques","score":0.9977999925613403,"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":0.8229782581329346},{"id":"https://openalex.org/keywords/security-token","display_name":"Security token","score":0.8222472667694092},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.5916121602058411},{"id":"https://openalex.org/keywords/transformer","display_name":"Transformer","score":0.5631369948387146},{"id":"https://openalex.org/keywords/discriminative-model","display_name":"Discriminative model","score":0.5280312299728394},{"id":"https://openalex.org/keywords/smoothing","display_name":"Smoothing","score":0.5219019055366516},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.43051955103874207},{"id":"https://openalex.org/keywords/computer-vision","display_name":"Computer vision","score":0.21916645765304565},{"id":"https://openalex.org/keywords/voltage","display_name":"Voltage","score":0.09425690770149231}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8229782581329346},{"id":"https://openalex.org/C48145219","wikidata":"https://www.wikidata.org/wiki/Q1335365","display_name":"Security token","level":2,"score":0.8222472667694092},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5916121602058411},{"id":"https://openalex.org/C66322947","wikidata":"https://www.wikidata.org/wiki/Q11658","display_name":"Transformer","level":3,"score":0.5631369948387146},{"id":"https://openalex.org/C97931131","wikidata":"https://www.wikidata.org/wiki/Q5282087","display_name":"Discriminative model","level":2,"score":0.5280312299728394},{"id":"https://openalex.org/C3770464","wikidata":"https://www.wikidata.org/wiki/Q775963","display_name":"Smoothing","level":2,"score":0.5219019055366516},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.43051955103874207},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.21916645765304565},{"id":"https://openalex.org/C165801399","wikidata":"https://www.wikidata.org/wiki/Q25428","display_name":"Voltage","level":2,"score":0.09425690770149231},{"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/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","level":0,"score":0.0},{"id":"https://openalex.org/C38652104","wikidata":"https://www.wikidata.org/wiki/Q3510521","display_name":"Computer security","level":1,"score":0.0}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1109/tpami.2022.3208922","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tpami.2022.3208922","pdf_url":null,"source":{"id":"https://openalex.org/S199944782","display_name":"IEEE Transactions on Pattern Analysis and Machine Intelligence","issn_l":"0162-8828","issn":["0162-8828","1939-3539","2160-9292"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310320439","host_organization_name":"IEEE Computer Society","host_organization_lineage":["https://openalex.org/P4310320439","https://openalex.org/P4310319808"],"host_organization_lineage_names":["IEEE Computer Society","Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Transactions on Pattern Analysis and Machine Intelligence","raw_type":"journal-article"},{"id":"pmid:36155475","is_oa":false,"landing_page_url":"https://pubmed.ncbi.nlm.nih.gov/36155475","pdf_url":null,"source":{"id":"https://openalex.org/S4306525036","display_name":"PubMed","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I1299303238","host_organization_name":"National Institutes of Health","host_organization_lineage":["https://openalex.org/I1299303238"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE transactions on pattern analysis and machine intelligence","raw_type":null}],"best_oa_location":null,"sustainable_development_goals":[{"display_name":"Reduced inequalities","id":"https://metadata.un.org/sdg/10","score":0.7400000095367432}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":92,"referenced_works":["https://openalex.org/W2108598243","https://openalex.org/W2163605009","https://openalex.org/W2194775991","https://openalex.org/W2293634267","https://openalex.org/W2302255633","https://openalex.org/W2407776548","https://openalex.org/W2560023338","https://openalex.org/W2765407302","https://openalex.org/W2884822772","https://openalex.org/W2908510526","https://openalex.org/W2923014074","https://openalex.org/W2963163009","https://openalex.org/W2963263347","https://openalex.org/W2963351113","https://openalex.org/W2964309882","https://openalex.org/W3034502973","https://openalex.org/W3035022492","https://openalex.org/W3035682985","https://openalex.org/W3092462694","https://openalex.org/W3094502228","https://openalex.org/W3095422700","https://openalex.org/W3096609285","https://openalex.org/W3101415077","https://openalex.org/W3117450517","https://openalex.org/W3121523901","https://openalex.org/W3131500599","https://openalex.org/W3133696297","https://openalex.org/W3137278571","https://openalex.org/W3137963805","https://openalex.org/W3138516171","https://openalex.org/W3151130473","https://openalex.org/W3153465022","https://openalex.org/W3153842237","https://openalex.org/W3156811085","https://openalex.org/W3157528469","https://openalex.org/W3160694286","https://openalex.org/W3165924482","https://openalex.org/W3170874841","https://openalex.org/W3171125843","https://openalex.org/W3171516518","https://openalex.org/W3172509117","https://openalex.org/W3175515048","https://openalex.org/W3176153963","https://openalex.org/W4206706211","https://openalex.org/W4214493665","https://openalex.org/W4214634256","https://openalex.org/W4226272139","https://openalex.org/W4287203292","https://openalex.org/W4287824654","https://openalex.org/W4292779060","https://openalex.org/W4295312788","https://openalex.org/W4297775537","https://openalex.org/W4298395628","https://openalex.org/W4312443924","https://openalex.org/W4313007769","https://openalex.org/W4385245566","https://openalex.org/W6684191040","https://openalex.org/W6713582272","https://openalex.org/W6726497184","https://openalex.org/W6737664043","https://openalex.org/W6739901393","https://openalex.org/W6745136726","https://openalex.org/W6748481559","https://openalex.org/W6750615492","https://openalex.org/W6753421600","https://openalex.org/W6757817989","https://openalex.org/W6763367864","https://openalex.org/W6766978945","https://openalex.org/W6771917389","https://openalex.org/W6778485988","https://openalex.org/W6778883912","https://openalex.org/W6779602356","https://openalex.org/W6784094891","https://openalex.org/W6784333009","https://openalex.org/W6785665406","https://openalex.org/W6786026227","https://openalex.org/W6786585107","https://openalex.org/W6786708909","https://openalex.org/W6788071488","https://openalex.org/W6788135285","https://openalex.org/W6788305448","https://openalex.org/W6788556936","https://openalex.org/W6790690058","https://openalex.org/W6791705549","https://openalex.org/W6793632291","https://openalex.org/W6793697131","https://openalex.org/W6794345597","https://openalex.org/W6794559225","https://openalex.org/W6794906783","https://openalex.org/W6797360341","https://openalex.org/W6798016242","https://openalex.org/W6810859482"],"related_works":["https://openalex.org/W2983744209","https://openalex.org/W2375389409","https://openalex.org/W2961085424","https://openalex.org/W2026121273","https://openalex.org/W2801772698","https://openalex.org/W66955737","https://openalex.org/W2102106825","https://openalex.org/W4306674287","https://openalex.org/W3139434170","https://openalex.org/W4303201644"],"abstract_inverted_index":{"Vision":[0],"transformers":[1],"have":[2],"recently":[3],"attained":[4],"state-of-the-art":[5],"results":[6],"in":[7,108,125],"visual":[8],"recognition":[9],"tasks.":[10,210],"Their":[11],"success":[12],"is":[13,212],"largely":[14],"attributed":[15],"to":[16,42,53,68,72,100,118,152,167,176,185,193,199],"the":[17,22,26,75,78,84,170,187],"self-attention":[18,37,51,64,89,124],"component,":[19],"which":[20,103],"models":[21,198],"global":[23],"dependencies":[24],"among":[25],"image":[27,204],"patches":[28],"(tokens)":[29],"and":[30,90,121,143,208],"aggregates":[31],"them":[32],"into":[33],"higher-level":[34],"features.":[35,112],"However,":[36],"brings":[38],"significant":[39],"training":[40],"difficulties":[41],"ViTs.":[43],"Many":[44],"recent":[45],"works":[46],"thus":[47],"develop":[48,115],"various":[49,140],"new":[50],"components":[52],"alleviate":[54],"this":[55,58],"issue.":[56],"In":[57,178],"article,":[59],"instead":[60],"of":[61,77,88,196],"developing":[62],"complicated":[63],"mechanism,":[65],"we":[66,180],"aim":[67],"explore":[69],"simple":[70,116],"approaches":[71,117],"fully":[73],"release":[74],"potential":[76],"vanilla":[79],"self-attention.":[80],"We":[81,113],"first":[82],"study":[83],"token":[85,111,126,130,188],"selection":[86],"behavior":[87],"find":[91],"that":[92],"it":[93],"suffers":[94],"from":[95],"a":[96,136,194],"low":[97],"diversity":[98,122],"due":[99],"attention":[101],"over-smoothing,":[102],"severely":[104],"limits":[105],"its":[106],"effectiveness":[107],"learning":[109],"discriminative":[110],"then":[114],"enhance":[119],"selectivity":[120],"for":[123,139,203],"selection.":[127],"The":[128],"resulted":[129],"selector":[131,189],"module":[132,138],"can":[133,172,190],"server":[134],"as":[135],"drop-in":[137],"ViT":[141],"backbones":[142],"consistently":[144],"boost":[145,200],"their":[146,201],"performance.":[147],"Significantly,":[148],"they":[149],"enable":[150],"ViTs":[151],"achieve":[153],"84.6%":[154],"top-1":[155],"classification":[156],"accuracy":[157],"on":[158],"ImageNet":[159],"with":[160],"only":[161],"25M":[162],"parameters.":[163],"When":[164],"scaled":[165],"up":[166],"81M":[168],"parameters,":[169],"result":[171],"be":[173,191],"further":[174],"improved":[175],"86.1%.":[177],"addition,":[179],"also":[181],"present":[182],"comprehensive":[183],"experiments":[184],"demonstrate":[186],"applied":[192],"variety":[195],"transformer-based":[197],"performance":[202],"classification,":[205],"semantic":[206],"segmentation":[207],"NLP":[209],"Code":[211],"available":[213],"at":[214],"https://github.com/zhoudaquan/dvit_repo.":[215]},"counts_by_year":[{"year":2025,"cited_by_count":7},{"year":2024,"cited_by_count":2},{"year":2023,"cited_by_count":4}],"updated_date":"2025-11-06T03:46:38.306776","created_date":"2025-10-10T00:00:00"}