{"id":"https://openalex.org/W3143320354","doi":"https://doi.org/10.1109/iccv48922.2021.00147","title":"UniT: Multimodal Multitask Learning with a Unified Transformer","display_name":"UniT: Multimodal Multitask Learning with a Unified Transformer","publication_year":2021,"publication_date":"2021-10-01","ids":{"openalex":"https://openalex.org/W3143320354","doi":"https://doi.org/10.1109/iccv48922.2021.00147","mag":"3143320354"},"language":"en","primary_location":{"id":"doi:10.1109/iccv48922.2021.00147","doi":"https://doi.org/10.1109/iccv48922.2021.00147","is_oa":false,"landing_page_url":"https://doi.org/10.1109/iccv48922.2021.00147","pdf_url":null,"source":{"id":"https://openalex.org/S4363607764","display_name":"IEEE/CVF International Conference on Computer Vision (ICCV)","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":"2021 IEEE/CVF International Conference on Computer Vision (ICCV)","raw_type":"proceedings-article"},"volume":null,"type":"conference-paper","is_traditional":true,"indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/abs/2102.10772v1","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5072304825","display_name":"Ronghang Hu","orcid":"https://orcid.org/0000-0002-5060-9485","observed_orcids":["https://orcid.org/0000-0002-5060-9485"]},"institutions":[{"id":"https://openalex.org/I4210114444","display_name":"Meta (United States)","ror":"https://ror.org/01zbnvs85","country_code":"US","type":"company","lineage":["https://openalex.org/I4210114444"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Ronghang Hu","raw_affiliation_strings":["Facebook AI Research (FAIR)"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Facebook AI Research (FAIR)","institution_ids":["https://openalex.org/I4210114444"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5100672985","display_name":"Amanpreet Singh","orcid":"https://orcid.org/0000-0002-3409-9833","observed_orcids":["https://orcid.org/0000-0002-3409-9833"]},"institutions":[{"id":"https://openalex.org/I4210114444","display_name":"Meta (United States)","ror":"https://ror.org/01zbnvs85","country_code":"US","type":"company","lineage":["https://openalex.org/I4210114444"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Amanpreet Singh","raw_affiliation_strings":["Facebook AI Research (FAIR)"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Facebook AI Research (FAIR)","institution_ids":["https://openalex.org/I4210114444"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I4210114444"],"apc_list":null,"apc_paid":null,"fwci":19.4545,"has_fulltext":false,"cited_by_count":335,"citation_normalized_percentile":{"value":0.99463441,"is_in_top_1_percent":true,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":99,"max":100},"biblio":{"volume":null,"issue":null,"first_page":"1419","last_page":"1429"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11714","display_name":"Multimodal Machine Learning 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/T11714","display_name":"Multimodal Machine Learning 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/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.001500000013038516,"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/T10036","display_name":"Advanced Neural Network Applications","score":0.0003000000142492354,"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/object-detection","display_name":"object detection","score":0.5550000071525574},{"id":"https://openalex.org/keywords/natural-language-understanding","display_name":"natural language understanding","score":0.9950000047683716},{"id":"https://openalex.org/keywords/multimodal-reasoning","display_name":"multimodal reasoning","score":0.9950000047683716},{"id":"https://openalex.org/keywords/parameter-efficiency","display_name":"parameter efficiency","score":0.5569999814033508}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8100907206535339},{"id":"https://openalex.org/C66322947","wikidata":"https://www.wikidata.org/wiki/Q11658","display_name":"Transformer","level":3,"score":0.7537931799888611},{"id":"https://openalex.org/C118505674","wikidata":"https://www.wikidata.org/wiki/Q42586063","display_name":"Encoder","level":2,"score":0.7092152237892151},{"id":"https://openalex.org/C28006648","wikidata":"https://www.wikidata.org/wiki/Q6934509","display_name":"Multi-task learning","level":3,"score":0.5605794191360474},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.49644285440444946},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.4695577919483185},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.43745407462120056},{"id":"https://openalex.org/C175154964","wikidata":"https://www.wikidata.org/wiki/Q380077","display_name":"Task analysis","level":3,"score":0.4239940941333771},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.4093881845474243},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.3275248408317566},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.08022135496139526},{"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/C201995342","wikidata":"https://www.wikidata.org/wiki/Q682496","display_name":"Systems engineering","level":1,"score":0.0},{"id":"https://openalex.org/C165801399","wikidata":"https://www.wikidata.org/wiki/Q25428","display_name":"Voltage","level":2,"score":0.0},{"id":"https://openalex.org/C119599485","wikidata":"https://www.wikidata.org/wiki/Q43035","display_name":"Electrical engineering","level":1,"score":0.0}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1109/iccv48922.2021.00147","doi":"https://doi.org/10.1109/iccv48922.2021.00147","is_oa":false,"landing_page_url":"https://doi.org/10.1109/iccv48922.2021.00147","pdf_url":null,"source":{"id":"https://openalex.org/S4363607764","display_name":"IEEE/CVF International Conference on Computer Vision (ICCV)","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":"2021 IEEE/CVF International Conference on Computer Vision (ICCV)","raw_type":"proceedings-article"},{"id":"mag:3129436779","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2102.10772v1","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":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"arXiv (Cornell University)","raw_type":null}],"best_oa_location":{"id":"mag:3129436779","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2102.10772v1","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":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"arXiv (Cornell University)","raw_type":null},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":71,"referenced_works":["https://openalex.org/W639708223","https://openalex.org/W1522301498","https://openalex.org/W1861492603","https://openalex.org/W2194775991","https://openalex.org/W2251939518","https://openalex.org/W2277195237","https://openalex.org/W2286929393","https://openalex.org/W2462831000","https://openalex.org/W2516255829","https://openalex.org/W2518108298","https://openalex.org/W2556468274","https://openalex.org/W2560730294","https://openalex.org/W2607892599","https://openalex.org/W2626778328","https://openalex.org/W2626792426","https://openalex.org/W2738574128","https://openalex.org/W2745461083","https://openalex.org/W2798512429","https://openalex.org/W2896457183","https://openalex.org/W2898732869","https://openalex.org/W2908510526","https://openalex.org/W2912371042","https://openalex.org/W2913340405","https://openalex.org/W2913568332","https://openalex.org/W2922913792","https://openalex.org/W2923014074","https://openalex.org/W2937843571","https://openalex.org/W2942489928","https://openalex.org/W2949718784","https://openalex.org/W2950813464","https://openalex.org/W2952603081","https://openalex.org/W2961117861","https://openalex.org/W2962897020","https://openalex.org/W2963091558","https://openalex.org/W2963150697","https://openalex.org/W2963168538","https://openalex.org/W2963310665","https://openalex.org/W2963341956","https://openalex.org/W2963748441","https://openalex.org/W2963854351","https://openalex.org/W2964121744","https://openalex.org/W2965373594","https://openalex.org/W2968124245","https://openalex.org/W2969876226","https://openalex.org/W2970231061","https://openalex.org/W2970608575","https://openalex.org/W2970971581","https://openalex.org/W2975059944","https://openalex.org/W2981852735","https://openalex.org/W2989226908","https://openalex.org/W2994759459","https://openalex.org/W2995460200","https://openalex.org/W2997591391","https://openalex.org/W2998356391","https://openalex.org/W3030163527","https://openalex.org/W3030520226","https://openalex.org/W3033102790","https://openalex.org/W3034336960","https://openalex.org/W3034445277","https://openalex.org/W3034588855","https://openalex.org/W3034672970","https://openalex.org/W3034727271","https://openalex.org/W3034885317","https://openalex.org/W3082274269","https://openalex.org/W3087549734","https://openalex.org/W3090449556","https://openalex.org/W3092462694","https://openalex.org/W3094502228","https://openalex.org/W3101065397","https://openalex.org/W3114896399","https://openalex.org/W3170874841"],"related_works":["https://openalex.org/W3129436779","https://openalex.org/W4385004765","https://openalex.org/W4402754178","https://openalex.org/W4312836739","https://openalex.org/W4280547190","https://openalex.org/W4313166855","https://openalex.org/W4283208789","https://openalex.org/W4390873285","https://openalex.org/W4366457978","https://openalex.org/W4390874224"],"abstract_inverted_index":{"We":[0],"propose":[1],"UniT,":[2],"a":[3,51,102],"Unified":[4],"Transformer":[5],"model":[6,36,66,89],"to":[7,21,77],"simultaneously":[8],"learn":[9,115],"the":[10,30,55,87],"most":[11],"prominent":[12],"tasks":[13,93,107,117],"across":[14,91,108],"different":[15,109],"domains,":[16],"ranging":[17],"from":[18,73],"object":[19],"detection":[20],"natural":[22],"language":[23],"understanding":[24],"and":[25,44,100],"multimodal":[26],"reasoning.":[27],"Based":[28],"on":[29,47,80,125],"transformer":[31],"encoder-decoder":[32],"architecture,":[33],"our":[34,112],"UniT":[35],"encodes":[37],"each":[38,48,74,126],"input":[39,57],"modality":[40],"with":[41,50,71,83,128],"an":[42],"encoder":[43],"makes":[45],"predictions":[46],"task":[49,127],"shared":[52],"decoder":[53],"over":[54,119],"encoded":[56],"representations,":[58],"followed":[59],"by":[60],"task-specific":[61,98],"output":[62],"heads.":[63],"The":[64],"entire":[65],"is":[67,134],"jointly":[68,118],"trained":[69],"end-to-end":[70],"losses":[72],"task.":[75],"Compared":[76],"previous":[78],"efforts":[79],"multi-task":[81],"learning":[82],"transformers,":[84],"we":[85,114],"share":[86],"same":[88],"parameters":[90],"all":[92],"instead":[94],"of":[95,106],"separately":[96],"fine-tuning":[97],"models":[99],"handle":[101],"much":[103],"higher":[104],"variety":[105],"domains.":[110],"In":[111],"experiments,":[113],"7":[116],"8":[120],"datasets,":[121],"achieving":[122],"strong":[123],"performance":[124],"significantly":[129],"fewer":[130],"parameters.":[131],"Our":[132],"code":[133],"available":[135],"in":[136],"MMF":[137],"at":[138],"https://mmf.sh.":[139]},"counts_by_year":[{"year":2026,"cited_by_count":29},{"year":2025,"cited_by_count":55},{"year":2024,"cited_by_count":81},{"year":2023,"cited_by_count":89},{"year":2022,"cited_by_count":60},{"year":2021,"cited_by_count":21}],"updated_date":"2026-11-07T11:25:07.637180","created_date":"2021-04-13T00:00:00"}