{"id":"https://openalex.org/W7155363338","doi":"https://doi.org/10.48550/arxiv.2604.20306","title":"Dual Causal Inference: Integrating Backdoor Adjustment and Instrumental Variable Learning for Medical VQA","display_name":"Dual Causal Inference: Integrating Backdoor Adjustment and Instrumental Variable Learning for Medical VQA","publication_year":2026,"publication_date":"2026-04-22","ids":{"openalex":"https://openalex.org/W7155363338","doi":"https://doi.org/10.48550/arxiv.2604.20306"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2604.20306","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.20306","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"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":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2604.20306","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5134372175","display_name":"Zibo Xu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xu, Zibo","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5134447078","display_name":"Qiang Li","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Li, Qiang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5134456967","display_name":"Ke Lu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Lu, Ke","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5134419081","display_name":"Jin Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Jin","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5134455013","display_name":"Weizhi Nie","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Nie, Weizhi","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5130787576","display_name":"Yuting Su","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Su, Yuting","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":null,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"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/T11714","display_name":"Multimodal Machine Learning Applications","score":0.9900000095367432,"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":0.9900000095367432,"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":0.003599999938160181,"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/T10028","display_name":"Topic Modeling","score":0.0007999999797903001,"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/interpretability","display_name":"Interpretability","score":0.765500009059906},{"id":"https://openalex.org/keywords/spurious-relationship","display_name":"Spurious relationship","score":0.6647999882698059},{"id":"https://openalex.org/keywords/instrumental-variable","display_name":"Instrumental variable","score":0.6442000269889832},{"id":"https://openalex.org/keywords/causal-inference","display_name":"Causal inference","score":0.6261000037193298},{"id":"https://openalex.org/keywords/backdoor","display_name":"Backdoor","score":0.6133000254631042},{"id":"https://openalex.org/keywords/robustness","display_name":"Robustness (evolution)","score":0.5778999924659729},{"id":"https://openalex.org/keywords/causal-model","display_name":"Causal model","score":0.5354999899864197},{"id":"https://openalex.org/keywords/endogeneity","display_name":"Endogeneity","score":0.5241000056266785},{"id":"https://openalex.org/keywords/overfitting","display_name":"Overfitting","score":0.5198000073432922}],"concepts":[{"id":"https://openalex.org/C2781067378","wikidata":"https://www.wikidata.org/wiki/Q17027399","display_name":"Interpretability","level":2,"score":0.765500009059906},{"id":"https://openalex.org/C97256817","wikidata":"https://www.wikidata.org/wiki/Q1462316","display_name":"Spurious relationship","level":2,"score":0.6647999882698059},{"id":"https://openalex.org/C162144332","wikidata":"https://www.wikidata.org/wiki/Q1665305","display_name":"Instrumental variable","level":2,"score":0.6442000269889832},{"id":"https://openalex.org/C158600405","wikidata":"https://www.wikidata.org/wiki/Q5054566","display_name":"Causal inference","level":2,"score":0.6261000037193298},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.6180999875068665},{"id":"https://openalex.org/C2781045450","wikidata":"https://www.wikidata.org/wiki/Q254569","display_name":"Backdoor","level":2,"score":0.6133000254631042},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6033999919891357},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5982000231742859},{"id":"https://openalex.org/C63479239","wikidata":"https://www.wikidata.org/wiki/Q7353546","display_name":"Robustness (evolution)","level":3,"score":0.5778999924659729},{"id":"https://openalex.org/C11671645","wikidata":"https://www.wikidata.org/wiki/Q5054567","display_name":"Causal model","level":2,"score":0.5354999899864197},{"id":"https://openalex.org/C610760","wikidata":"https://www.wikidata.org/wiki/Q1340706","display_name":"Endogeneity","level":2,"score":0.5241000056266785},{"id":"https://openalex.org/C22019652","wikidata":"https://www.wikidata.org/wiki/Q331309","display_name":"Overfitting","level":3,"score":0.5198000073432922},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.517799973487854},{"id":"https://openalex.org/C149782125","wikidata":"https://www.wikidata.org/wiki/Q160039","display_name":"Econometrics","level":1,"score":0.42800000309944153},{"id":"https://openalex.org/C51167844","wikidata":"https://www.wikidata.org/wiki/Q4422623","display_name":"Latent variable","level":2,"score":0.42570000886917114},{"id":"https://openalex.org/C2780980858","wikidata":"https://www.wikidata.org/wiki/Q110022","display_name":"Dual (grammatical number)","level":2,"score":0.4059999883174896},{"id":"https://openalex.org/C182365436","wikidata":"https://www.wikidata.org/wiki/Q50701","display_name":"Variable (mathematics)","level":2,"score":0.40049999952316284},{"id":"https://openalex.org/C77350462","wikidata":"https://www.wikidata.org/wiki/Q1125472","display_name":"Confounding","level":2,"score":0.3474999964237213},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.34369999170303345},{"id":"https://openalex.org/C34736171","wikidata":"https://www.wikidata.org/wiki/Q918333","display_name":"Preprocessor","level":2,"score":0.33959999680519104},{"id":"https://openalex.org/C64357122","wikidata":"https://www.wikidata.org/wiki/Q1149766","display_name":"Causality (physics)","level":2,"score":0.33880001306533813},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.3077999949455261},{"id":"https://openalex.org/C79337645","wikidata":"https://www.wikidata.org/wiki/Q779824","display_name":"Outlier","level":2,"score":0.30140000581741333},{"id":"https://openalex.org/C108154423","wikidata":"https://www.wikidata.org/wiki/Q1469792","display_name":"Salience (neuroscience)","level":2,"score":0.28529998660087585},{"id":"https://openalex.org/C59404180","wikidata":"https://www.wikidata.org/wiki/Q17013334","display_name":"Feature learning","level":2,"score":0.2782000005245209},{"id":"https://openalex.org/C184337299","wikidata":"https://www.wikidata.org/wiki/Q1437428","display_name":"Semantics (computer science)","level":2,"score":0.27630001306533813},{"id":"https://openalex.org/C28006648","wikidata":"https://www.wikidata.org/wiki/Q6934509","display_name":"Multi-task learning","level":3,"score":0.27320000529289246},{"id":"https://openalex.org/C32848918","wikidata":"https://www.wikidata.org/wiki/Q845789","display_name":"Observable","level":2,"score":0.26260000467300415},{"id":"https://openalex.org/C194541083","wikidata":"https://www.wikidata.org/wiki/Q457174","display_name":"Workaround","level":2,"score":0.25769999623298645},{"id":"https://openalex.org/C2780695315","wikidata":"https://www.wikidata.org/wiki/Q3799040","display_name":"Unobservable","level":2,"score":0.2556999921798706},{"id":"https://openalex.org/C163504300","wikidata":"https://www.wikidata.org/wiki/Q2364925","display_name":"Causal structure","level":2,"score":0.2529999911785126}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2604.20306","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.20306","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"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":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2604.20306","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.20306","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"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":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Medical":[0],"Visual":[1],"Question":[2],"Answering":[3],"(MedVQA)":[4],"aims":[5],"to":[6,23,40,48,90],"generate":[7],"clinically":[8],"reliable":[9],"answers":[10],"conditioned":[11],"on":[12,150,181],"complex":[13],"medical":[14,34],"images":[15],"and":[16,85,95,113,163,188,213],"questions.":[17],"However,":[18],"existing":[19,196],"methods":[20],"often":[21],"overfit":[22],"superficial":[24],"cross-modal":[25,41,108,216,226],"correlations,":[26],"neglecting":[27],"the":[28,69,76,136,139,151,160,211],"intrinsic":[29],"biases":[30,109],"embedded":[31],"in":[32,199],"multimodal":[33,153],"data.":[35],"Consequently,":[36],"models":[37],"become":[38],"vulnerable":[39],"confounding":[42],"effects,":[43],"severely":[44],"hindering":[45],"their":[46],"ability":[47],"provide":[49],"trustworthy":[50],"diagnostic":[51],"reasoning.":[52],"To":[53,68,134],"address":[54],"this":[55,167],"limitation,":[56],"we":[57,99,141],"propose":[58],"a":[59,101,130],"novel":[60],"Dual":[61],"Causal":[62,103],"Inference":[63],"(DCI)":[64],"framework":[65],"for":[66],"MedVQA.":[67],"best":[70],"of":[71,138,215],"our":[72,192],"knowledge,":[73],"DCI":[74,170,208],"is":[75],"first":[77],"unified":[78],"architecture":[79],"that":[80,146,174,191,207],"integrates":[81],"Backdoor":[82],"Adjustment":[83],"(BDA)":[84],"Instrumental":[86],"Variable":[87],"(IV)":[88],"learning":[89],"jointly":[91],"tackle":[92],"both":[93],"observable":[94,107],"unobserved":[96,121,161],"confounders.":[97],"Specifically,":[98],"formulate":[100],"Structural":[102],"Model":[104],"(SCM)":[105],"where":[106],"(e.g.,":[110],"frequent":[111],"visual":[112],"textual":[114],"co-occurrences)":[115],"are":[116,123],"mitigated":[117],"via":[118],"BDA,":[119],"while":[120,155],"confounders":[122,162],"compensated":[124],"using":[125],"an":[126],"IV":[127],"learned":[128],"from":[129,224],"shared":[131],"latent":[132],"space.":[133],"guarantee":[135],"validity":[137],"IV,":[140],"design":[142],"mutual":[143],"information":[144],"constraints":[145],"maximize":[147],"its":[148,157],"dependence":[149],"fused":[152],"representations":[154,173],"minimizing":[156],"associations":[158],"with":[159],"target":[164],"answers.":[165],"Through":[166],"dual":[168],"mechanism,":[169],"extracts":[171],"deconfounded":[172],"capture":[175],"genuine":[176],"causal":[177,222],"relationships.":[178],"Extensive":[179],"experiments":[180],"four":[182],"benchmark":[183],"datasets,":[184],"SLAKE,":[185],"SLAKE-CP,":[186],"VQA-RAD,":[187],"PathVQA,":[189],"demonstrate":[190],"method":[193],"consistently":[194],"outperforms":[195],"approaches,":[197],"particularly":[198],"out-of-distribution":[200],"(OOD)":[201],"generalization.":[202],"Furthermore,":[203],"qualitative":[204],"analyses":[205],"confirm":[206],"significantly":[209],"enhances":[210],"interpretability":[212],"robustness":[214],"reasoning":[217],"by":[218],"explicitly":[219],"disentangling":[220],"true":[221],"effects":[223],"spurious":[225],"shortcuts.":[227]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-04-24T00:00:00"}