{"id":"https://openalex.org/W3206022579","doi":"https://doi.org/10.1109/tpami.2021.3121705","title":"Deconfounded Image Captioning: A Causal Retrospect","display_name":"Deconfounded Image Captioning: A Causal Retrospect","publication_year":2021,"publication_date":"2021-12-14","ids":{"openalex":"https://openalex.org/W3206022579","doi":"https://doi.org/10.1109/tpami.2021.3121705","mag":"3206022579","pmid":"https://pubmed.ncbi.nlm.nih.gov/34673483"},"language":"en","primary_location":{"id":"doi:10.1109/tpami.2021.3121705","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tpami.2021.3121705","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/A5064752243","display_name":"Xu Yang","orcid":"https://orcid.org/0000-0002-8276-2679"},"institutions":[{"id":"https://openalex.org/I172675005","display_name":"Nanyang Technological University","ror":"https://ror.org/02e7b5302","country_code":"SG","type":"education","lineage":["https://openalex.org/I172675005"]},{"id":"https://openalex.org/I4210090971","display_name":"Southeast University","ror":"https://ror.org/00cf0ab87","country_code":"BD","type":"education","lineage":["https://openalex.org/I4210090971"]},{"id":"https://openalex.org/I56590836","display_name":"Monash University","ror":"https://ror.org/02bfwt286","country_code":"AU","type":"education","lineage":["https://openalex.org/I56590836"]},{"id":"https://openalex.org/I76569877","display_name":"Southeast University","ror":"https://ror.org/04ct4d772","country_code":"CN","type":"education","lineage":["https://openalex.org/I76569877"]}],"countries":["AU","BD","CN","SG"],"is_corresponding":false,"raw_author_name":"Xu Yang","raw_affiliation_strings":["school of computer science, Nanyang Technological University, 54761 Singapore, singapore, Singapore, 639798 (e-mail: s170018@e.ntu.edu.sg)","Data Science & AI Department at the Faculty of IT, Monash University, Australia","School of Computer Science and Engineering of Southeast University, China","Nanyang Techno-logical University, Singapore"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"school of computer science, Nanyang Technological University, 54761 Singapore, singapore, Singapore, 639798 (e-mail: s170018@e.ntu.edu.sg)","institution_ids":["https://openalex.org/I172675005"]},{"raw_affiliation_string":"Data Science & AI Department at the Faculty of IT, Monash University, Australia","institution_ids":["https://openalex.org/I56590836"]},{"raw_affiliation_string":"School of Computer Science and Engineering of Southeast University, China","institution_ids":["https://openalex.org/I4210090971","https://openalex.org/I76569877"]},{"raw_affiliation_string":"Nanyang Techno-logical University, Singapore","institution_ids":["https://openalex.org/I172675005"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5042324027","display_name":"Hanwang Zhang","orcid":"https://orcid.org/0000-0001-7374-8739"},"institutions":[{"id":"https://openalex.org/I172675005","display_name":"Nanyang Technological University","ror":"https://ror.org/02e7b5302","country_code":"SG","type":"education","lineage":["https://openalex.org/I172675005"]},{"id":"https://openalex.org/I4210090971","display_name":"Southeast University","ror":"https://ror.org/00cf0ab87","country_code":"BD","type":"education","lineage":["https://openalex.org/I4210090971"]},{"id":"https://openalex.org/I56590836","display_name":"Monash University","ror":"https://ror.org/02bfwt286","country_code":"AU","type":"education","lineage":["https://openalex.org/I56590836"]},{"id":"https://openalex.org/I76569877","display_name":"Southeast University","ror":"https://ror.org/04ct4d772","country_code":"CN","type":"education","lineage":["https://openalex.org/I76569877"]}],"countries":["AU","BD","CN","SG"],"is_corresponding":false,"raw_author_name":"Hanwang Zhang","raw_affiliation_strings":["School of Computer Science and Engineering, Nanyang Technological University, 54761 Singapore, Singapore, Singapore, (e-mail: zhanghw.ntu@gmail.com)","School of Computer Science and Engineering of Southeast University, China","Nanyang Techno-logical University, Singapore","Data Science & AI Department at the Faculty of IT, Monash University, Australia"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Computer Science and Engineering, Nanyang Technological University, 54761 Singapore, Singapore, Singapore, (e-mail: zhanghw.ntu@gmail.com)","institution_ids":["https://openalex.org/I172675005"]},{"raw_affiliation_string":"School of Computer Science and Engineering of Southeast University, China","institution_ids":["https://openalex.org/I4210090971","https://openalex.org/I76569877"]},{"raw_affiliation_string":"Nanyang Techno-logical University, Singapore","institution_ids":["https://openalex.org/I172675005"]},{"raw_affiliation_string":"Data Science & AI Department at the Faculty of IT, Monash University, Australia","institution_ids":["https://openalex.org/I56590836"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5100635804","display_name":"Jianfei Cai","orcid":"https://orcid.org/0000-0002-9444-3763"},"institutions":[{"id":"https://openalex.org/I172675005","display_name":"Nanyang Technological University","ror":"https://ror.org/02e7b5302","country_code":"SG","type":"education","lineage":["https://openalex.org/I172675005"]},{"id":"https://openalex.org/I4210090971","display_name":"Southeast University","ror":"https://ror.org/00cf0ab87","country_code":"BD","type":"education","lineage":["https://openalex.org/I4210090971"]},{"id":"https://openalex.org/I56590836","display_name":"Monash University","ror":"https://ror.org/02bfwt286","country_code":"AU","type":"education","lineage":["https://openalex.org/I56590836"]},{"id":"https://openalex.org/I76569877","display_name":"Southeast University","ror":"https://ror.org/04ct4d772","country_code":"CN","type":"education","lineage":["https://openalex.org/I76569877"]}],"countries":["AU","BD","CN","SG"],"is_corresponding":false,"raw_author_name":"Jianfei Cai","raw_affiliation_strings":["Faculty of Information Technology, Monash University, 2541 Clayton, Victoria, Australia, (e-mail: jianfei.cai@monash.edu)","School of Computer Science and Engineering of Southeast University, China","Data Science & AI Department at the Faculty of IT, Monash University, Australia","Nanyang Techno-logical University, Singapore"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Faculty of Information Technology, Monash University, 2541 Clayton, Victoria, Australia, (e-mail: jianfei.cai@monash.edu)","institution_ids":["https://openalex.org/I56590836"]},{"raw_affiliation_string":"School of Computer Science and Engineering of Southeast University, China","institution_ids":["https://openalex.org/I4210090971","https://openalex.org/I76569877"]},{"raw_affiliation_string":"Data Science & AI Department at the Faculty of IT, Monash University, Australia","institution_ids":["https://openalex.org/I56590836"]},{"raw_affiliation_string":"Nanyang Techno-logical University, Singapore","institution_ids":["https://openalex.org/I172675005"]}]}],"institutions":[],"countries_distinct_count":4,"institutions_distinct_count":4,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":7.6841,"has_fulltext":false,"cited_by_count":124,"citation_normalized_percentile":{"value":0.98387429,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":97,"max":100},"biblio":{"volume":"45","issue":"11","first_page":"1","last_page":"1"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11714","display_name":"Multimodal Machine Learning Applications","score":1.0,"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":1.0,"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":0.9980999827384949,"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":0.9890999794006348,"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/closed-captioning","display_name":"Closed captioning","score":0.9693936109542847},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.7093116044998169},{"id":"https://openalex.org/keywords/image","display_name":"Image (mathematics)","score":0.6411125659942627},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.5841588377952576},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.5399749875068665},{"id":"https://openalex.org/keywords/perspective","display_name":"Perspective (graphical)","score":0.5384677648544312},{"id":"https://openalex.org/keywords/causal-inference","display_name":"Causal inference","score":0.4995424747467041},{"id":"https://openalex.org/keywords/natural-language","display_name":"Natural language","score":0.44177109003067017},{"id":"https://openalex.org/keywords/backdoor","display_name":"Backdoor","score":0.4298650026321411},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.4271157383918762},{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.38250043988227844},{"id":"https://openalex.org/keywords/econometrics","display_name":"Econometrics","score":0.14306244254112244},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.11191248893737793},{"id":"https://openalex.org/keywords/computer-security","display_name":"Computer security","score":0.11052528023719788}],"concepts":[{"id":"https://openalex.org/C157657479","wikidata":"https://www.wikidata.org/wiki/Q2367247","display_name":"Closed captioning","level":3,"score":0.9693936109542847},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7093116044998169},{"id":"https://openalex.org/C115961682","wikidata":"https://www.wikidata.org/wiki/Q860623","display_name":"Image (mathematics)","level":2,"score":0.6411125659942627},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5841588377952576},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.5399749875068665},{"id":"https://openalex.org/C12713177","wikidata":"https://www.wikidata.org/wiki/Q1900281","display_name":"Perspective (graphical)","level":2,"score":0.5384677648544312},{"id":"https://openalex.org/C158600405","wikidata":"https://www.wikidata.org/wiki/Q5054566","display_name":"Causal inference","level":2,"score":0.4995424747467041},{"id":"https://openalex.org/C195324797","wikidata":"https://www.wikidata.org/wiki/Q33742","display_name":"Natural language","level":2,"score":0.44177109003067017},{"id":"https://openalex.org/C2781045450","wikidata":"https://www.wikidata.org/wiki/Q254569","display_name":"Backdoor","level":2,"score":0.4298650026321411},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.4271157383918762},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.38250043988227844},{"id":"https://openalex.org/C149782125","wikidata":"https://www.wikidata.org/wiki/Q160039","display_name":"Econometrics","level":1,"score":0.14306244254112244},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.11191248893737793},{"id":"https://openalex.org/C38652104","wikidata":"https://www.wikidata.org/wiki/Q3510521","display_name":"Computer security","level":1,"score":0.11052528023719788}],"mesh":[],"locations_count":3,"locations":[{"id":"doi:10.1109/tpami.2021.3121705","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tpami.2021.3121705","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:34673483","is_oa":false,"landing_page_url":"https://pubmed.ncbi.nlm.nih.gov/34673483","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},{"id":"pmh:oai:dr.ntu.edu.sg:10356/162629","is_oa":false,"landing_page_url":"https://hdl.handle.net/10356/162629","pdf_url":null,"source":{"id":"https://openalex.org/S4306402609","display_name":"DR-NTU (Nanyang Technological University)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I172675005","host_organization_name":"Nanyang Technological University","host_organization_lineage":["https://openalex.org/I172675005"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Journal Article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":90,"referenced_works":["https://openalex.org/W1516659296","https://openalex.org/W1861492603","https://openalex.org/W1889081078","https://openalex.org/W1895577753","https://openalex.org/W1905882502","https://openalex.org/W1933349210","https://openalex.org/W1956340063","https://openalex.org/W1969616664","https://openalex.org/W1987835821","https://openalex.org/W2005876975","https://openalex.org/W2016089260","https://openalex.org/W2023060295","https://openalex.org/W2050482109","https://openalex.org/W2095705004","https://openalex.org/W2101105183","https://openalex.org/W2139906443","https://openalex.org/W2154652894","https://openalex.org/W2164370980","https://openalex.org/W2194775991","https://openalex.org/W2251512949","https://openalex.org/W2277195237","https://openalex.org/W2302086703","https://openalex.org/W2479423890","https://openalex.org/W2486285194","https://openalex.org/W2506483933","https://openalex.org/W2552161745","https://openalex.org/W2560730294","https://openalex.org/W2561715562","https://openalex.org/W2575842049","https://openalex.org/W2745461083","https://openalex.org/W2795151422","https://openalex.org/W2798959609","https://openalex.org/W2885013662","https://openalex.org/W2886641317","https://openalex.org/W2887585070","https://openalex.org/W2890531016","https://openalex.org/W2893425640","https://openalex.org/W2938603906","https://openalex.org/W2962735233","https://openalex.org/W2962749380","https://openalex.org/W2962779575","https://openalex.org/W2962933664","https://openalex.org/W2963084599","https://openalex.org/W2963101956","https://openalex.org/W2963109634","https://openalex.org/W2963349562","https://openalex.org/W2963448850","https://openalex.org/W2963518342","https://openalex.org/W2963536419","https://openalex.org/W2963644680","https://openalex.org/W2965697393","https://openalex.org/W2983141445","https://openalex.org/W2984121207","https://openalex.org/W2986670728","https://openalex.org/W2990069284","https://openalex.org/W2997591391","https://openalex.org/W3004349648","https://openalex.org/W3016970897","https://openalex.org/W3034655362","https://openalex.org/W3034682719","https://openalex.org/W3034984754","https://openalex.org/W3035017890","https://openalex.org/W3035103424","https://openalex.org/W3035145964","https://openalex.org/W3035323998","https://openalex.org/W3035651653","https://openalex.org/W3093017735","https://openalex.org/W3096612875","https://openalex.org/W3103934428","https://openalex.org/W3105136412","https://openalex.org/W3177934633","https://openalex.org/W4249013746","https://openalex.org/W4299515571","https://openalex.org/W6620707391","https://openalex.org/W6630875275","https://openalex.org/W6631190155","https://openalex.org/W6639432524","https://openalex.org/W6674330103","https://openalex.org/W6678262379","https://openalex.org/W6682631176","https://openalex.org/W6713582119","https://openalex.org/W6719270105","https://openalex.org/W6721933647","https://openalex.org/W6731370813","https://openalex.org/W6744097617","https://openalex.org/W6748581942","https://openalex.org/W6762931136","https://openalex.org/W6764756247","https://openalex.org/W6766904570","https://openalex.org/W6781965151"],"related_works":["https://openalex.org/W4320031223","https://openalex.org/W3015678314","https://openalex.org/W4281902577","https://openalex.org/W4200629851","https://openalex.org/W4386185023","https://openalex.org/W3009072493","https://openalex.org/W4317672133","https://openalex.org/W3140988292","https://openalex.org/W2803345361","https://openalex.org/W3009270862"],"abstract_inverted_index":{"Dataset":[0],"bias":[1],"in":[2],"vision-language":[3],"tasks":[4],"is":[5,78,139],"becoming":[6],"one":[7],"of":[8,16,52,130],"the":[9,14,50,70,86,131],"main":[10],"problems":[11],"which":[12,147],"hinders":[13],"progress":[15],"our":[17,144],"community.":[18],"Existing":[19],"solutions":[20],"lack":[21],"a":[22,40,64,116,140],"principled":[23],"analysis":[24],"about":[25],"why":[26],"modern":[27,57],"image":[28,59,152],"captioners":[29],"easily":[30],"collapse":[31],"into":[32],"dataset":[33,75],"bias.":[34,76],"In":[35,101],"this":[36,53],"paper,":[37],"we":[38,103],"present":[39],"novel":[41],"perspective:":[42],"Deconfounded":[43],"Image":[44],"Captioning":[45],"(DIC),":[46],"to":[47,68],"find":[48],"out":[49],"answer":[51],"question,":[54],"then":[55],"retrospect":[56],"neural":[58],"captioners,":[60],"and":[61,88,96,113,120,127],"finally":[62],"propose":[63],"DIC":[65,77],"framework:":[66],"DICv1.0":[67,106,138],"alleviate":[69],"negative":[71],"effects":[72],"brought":[73],"by":[74],"based":[79],"on":[80,124],"causal":[81,145],"inference,":[82],"whose":[83],"two":[84,109],"principles:":[85],"backdoor":[87],"front-door":[89],"adjustments,":[90],"help":[91],"us":[92],"review":[93],"previous":[94],"studies":[95],"design":[97],"new":[98],"effective":[99],"models.":[100],"particular,":[102],"showcase":[104],"that":[105],"can":[107,114],"strengthen":[108],"prevailing":[110],"captioning":[111],"models":[112],"achieve":[115],"single-model":[117],"131.1":[118],"CIDEr-D":[119,123],"128.4":[121],"c40":[122],"Karpathy":[125],"split":[126,129],"online":[128],"challenging":[132],"MS":[133],"COCO":[134],"dataset,":[135],"respectively.":[136],"Interestingly,":[137],"natural":[141],"derivation":[142],"from":[143],"retrospect,":[146],"opens":[148],"promising":[149],"directions":[150],"for":[151],"captioning.":[153]},"counts_by_year":[{"year":2026,"cited_by_count":4},{"year":2025,"cited_by_count":31},{"year":2024,"cited_by_count":26},{"year":2023,"cited_by_count":31},{"year":2022,"cited_by_count":21},{"year":2021,"cited_by_count":7},{"year":2020,"cited_by_count":4}],"updated_date":"2026-07-16T13:24:37.021932","created_date":"2025-10-10T00:00:00"}