{"id":"https://openalex.org/W4411244976","doi":"https://doi.org/10.5244/c.35.79","title":"Rethinking Token-Mixing MLP for MLP-based Vision Backbone","display_name":"Rethinking Token-Mixing MLP for MLP-based Vision Backbone","publication_year":2021,"publication_date":"2021-01-01","ids":{"openalex":"https://openalex.org/W4411244976","doi":"https://doi.org/10.5244/c.35.79"},"language":"en","primary_location":{"id":"doi:10.5244/c.35.79","doi":"https://doi.org/10.5244/c.35.79","is_oa":false,"landing_page_url":"https://doi.org/10.5244/c.35.79","pdf_url":null,"source":{"id":"https://openalex.org/S7407088578","display_name":"British Machine Vision Conference (BMVC)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the British Machine Vision Conference 2021","raw_type":"proceedings-article"},"volume":null,"type":"conference-paper","is_traditional":true,"indexed_in":["arxiv","crossref","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2106.14882","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5018887598","display_name":"Tan Yu","orcid":"https://orcid.org/0000-0001-6071-0395","observed_orcids":["https://orcid.org/0000-0001-6071-0395"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Tan Yu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100342455","display_name":"Xu Li","orcid":"https://orcid.org/0000-0002-5725-2677","observed_orcids":["https://orcid.org/0000-0002-5725-2677"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"XU LI","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5017058023","display_name":"Yunfeng Cai","orcid":"https://orcid.org/0000-0003-2387-191X","observed_orcids":["https://orcid.org/0000-0003-2387-191X"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yunfeng Cai","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5103150619","display_name":"Mingming Sun","orcid":"https://orcid.org/0000-0002-6199-4905","observed_orcids":["https://orcid.org/0000-0002-6199-4905"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Mingming Sun","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5100435527","display_name":"Ping Li","orcid":"https://orcid.org/0000-0002-8515-7773","observed_orcids":["https://orcid.org/0000-0002-8515-7773","https://orcid.org/0000-0002-5979-8868","https://orcid.org/0000-0002-7985-8477","https://orcid.org/0000-0002-1503-0240","https://orcid.org/0000-0002-8391-6510","https://orcid.org/0000-0001-8285-8921","https://orcid.org/0000-0003-4725-4553"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ping Li","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":1.4635,"has_fulltext":true,"cited_by_count":23,"citation_normalized_percentile":{"value":0.88220316,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":93,"max":98},"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":0.9922000169754028,"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":0.9922000169754028,"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":0.0035000001080334187,"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/T11714","display_name":"Multimodal Machine Learning Applications","score":0.0007999999797903001,"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/vision-transformer","display_name":"vision transformer","score":0.6190000176429749},{"id":"https://openalex.org/keywords/spatial-invariance","display_name":"spatial invariance","score":0.7329999804496765},{"id":"https://openalex.org/keywords/inductive-bias","display_name":"inductive bias","score":0.5889999866485596},{"id":"https://openalex.org/keywords/image-classification","display_name":"image classification","score":0.6470000147819519},{"id":"https://openalex.org/keywords/imagenet-1k","display_name":"ImageNet-1K","score":0.8330000042915344}],"concepts":[{"id":"https://openalex.org/C138777275","wikidata":"https://www.wikidata.org/wiki/Q6884054","display_name":"Mixing (physics)","level":2,"score":0.7453200221061707},{"id":"https://openalex.org/C48145219","wikidata":"https://www.wikidata.org/wiki/Q1335365","display_name":"Security token","level":2,"score":0.690963625907898},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6228131651878357},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5224096775054932},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.5037948489189148},{"id":"https://openalex.org/C31258907","wikidata":"https://www.wikidata.org/wiki/Q1301371","display_name":"Computer network","level":1,"score":0.11683037877082825},{"id":"https://openalex.org/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","level":0,"score":0.08486893773078918},{"id":"https://openalex.org/C62520636","wikidata":"https://www.wikidata.org/wiki/Q944","display_name":"Quantum mechanics","level":1,"score":0.0}],"mesh":[],"locations_count":3,"locations":[{"id":"doi:10.5244/c.35.79","doi":"https://doi.org/10.5244/c.35.79","is_oa":false,"landing_page_url":"https://doi.org/10.5244/c.35.79","pdf_url":null,"source":{"id":"https://openalex.org/S7407088578","display_name":"British Machine Vision Conference (BMVC)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the British Machine Vision Conference 2021","raw_type":"proceedings-article"},{"id":"pmh:oai:arXiv.org:2106.14882","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2106.14882","pdf_url":"https://arxiv.org/pdf/2106.14882","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"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":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2106.14882","doi":"https://doi.org/10.48550/arxiv.2106.14882","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2106.14882","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"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":false,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2106.14882","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2106.14882","pdf_url":"https://arxiv.org/pdf/2106.14882","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"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":null,"raw_type":"text"},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4411244976.pdf","grobid_xml":"https://content.openalex.org/works/W4411244976.grobid-xml"},"referenced_works_count":39,"referenced_works":["https://openalex.org/W2108598243","https://openalex.org/W2194775991","https://openalex.org/W2462831000","https://openalex.org/W2531409750","https://openalex.org/W2612445135","https://openalex.org/W2625477277","https://openalex.org/W2626778328","https://openalex.org/W2765407302","https://openalex.org/W2899663614","https://openalex.org/W2908510526","https://openalex.org/W2946948417","https://openalex.org/W2982091175","https://openalex.org/W3016265891","https://openalex.org/W3037932933","https://openalex.org/W3102631365","https://openalex.org/W3116489684","https://openalex.org/W3121523901","https://openalex.org/W3130071011","https://openalex.org/W3131500599","https://openalex.org/W3133696297","https://openalex.org/W3138516171","https://openalex.org/W3139049060","https://openalex.org/W3146097248","https://openalex.org/W3157506437","https://openalex.org/W3157528469","https://openalex.org/W3158846111","https://openalex.org/W3163465952","https://openalex.org/W3164540605","https://openalex.org/W3165088525","https://openalex.org/W3168101492","https://openalex.org/W3169769781","https://openalex.org/W3170974475","https://openalex.org/W3202715235","https://openalex.org/W3203606893","https://openalex.org/W3204076343","https://openalex.org/W3211432419","https://openalex.org/W4212819272","https://openalex.org/W4283809036","https://openalex.org/W4302275239"],"related_works":["https://openalex.org/W4399766323","https://openalex.org/W4386228587","https://openalex.org/W4212819272","https://openalex.org/W4221154952","https://openalex.org/W4313056180","https://openalex.org/W4389988508","https://openalex.org/W3157506437","https://openalex.org/W4382465803","https://openalex.org/W4406160536","https://openalex.org/W4386065534"],"abstract_inverted_index":{"In":[0,44,122],"the":[1,10,16,20,49,86,91,97,113,127,130,153,157],"past":[2],"decade,":[3],"we":[4,125,171],"have":[5,68],"witnessed":[6],"rapid":[7],"progress":[8],"in":[9,31,55,72,138,164],"machine":[11],"vision":[12,34,58,74,92],"backbone.":[13,43],"By":[14],"introducing":[15],"inductive":[17,65,98],"bias":[18],"from":[19],"image":[21],"processing,":[22],"convolution":[23],"neural":[24],"network":[25],"(CNN)":[26],"has":[27,37],"achieved":[28,52,69],"excellent":[29],"performance":[30,71],"numerous":[32],"computer":[33,73],"tasks":[35,75],"and":[36,115,144,186],"been":[38],"established":[39],"as":[40,177],"\\emph{de":[41],"facto}":[42],"recent":[45],"years,":[46],"inspired":[47],"by":[48,53],"great":[50],"success":[51],"Transformer":[54,59],"NLP":[56],"tasks,":[57],"models":[60],"emerge.":[61],"Using":[62],"much":[63],"less":[64],"bias,":[66,99],"they":[67],"promising":[70],"compared":[76],"with":[77],"their":[78,162],"CNN":[79],"counterparts.":[80],"More":[81],"recently,":[82],"researchers":[83],"investigate":[84],"using":[85],"pure-MLP":[87,104],"architecture":[88],"to":[89,94,111,149],"build":[90],"backbone":[93,105],"further":[95],"reduce":[96],"achieving":[100],"good":[101],"performance.":[102],"The":[103],"is":[106,147,184],"built":[107],"upon":[108],"channel-mixing":[109],"MLPs":[110,117,137,160],"fuse":[112],"channels":[114],"token-mixing":[116,131,136,159,181],"for":[118],"communications":[119],"between":[120],"patches.":[121],"this":[123],"paper,":[124],"re-think":[126],"design":[128],"of":[129,156],"MLP.":[132],"We":[133],"discover":[134],"that":[135],"existing":[139,158],"MLP-based":[140],"backbones":[141],"are":[142],"spatial-specific,":[143],"thus":[145],"it":[146],"sensitive":[148],"spatial":[150],"translation.":[151],"Meanwhile,":[152],"channel-agnostic":[154],"property":[155],"limits":[161],"capability":[163],"mixing":[165],"tokens.":[166],"To":[167],"overcome":[168],"those":[169],"limitations,":[170],"propose":[172],"an":[173],"improved":[174],"structure":[175],"termed":[176],"Circulant":[178],"Channel-Specific":[179],"(CCS)":[180],"MLP,":[182],"which":[183],"spatial-invariant":[185],"channel-specific.":[187],"It":[188],"takes":[189],"fewer":[190],"parameters":[191],"but":[192],"achieves":[193],"higher":[194],"classification":[195],"accuracy":[196],"on":[197],"ImageNet1K":[198],"benchmark.":[199]},"counts_by_year":[{"year":2026,"cited_by_count":2},{"year":2025,"cited_by_count":2},{"year":2024,"cited_by_count":6},{"year":2023,"cited_by_count":6},{"year":2022,"cited_by_count":5},{"year":2021,"cited_by_count":2}],"updated_date":"2026-11-10T11:39:43.021295","created_date":"2025-10-10T00:00:00"}