Matches in SemOpenAlex for { <https://semopenalex.org/work/W4366388861> ?p ?o ?g. }
Showing items 1 to 87 of
87
with 100 items per page.
- W4366388861 abstract "Existing visual question reasoning methods usually fail to explicitly discover the inherent causal mechanism and ignore the complex event-level understanding that requires jointly modeling cross-modal event temporality and causality. In this paper, we propose an event-level visual question reasoning framework named Cross-Modal Question Reasoning (CMQR), to explicitly discover temporal causal structure and mitigate visual spurious correlation by causal intervention. To explicitly discover visual causal structure, the Visual Causality Discovery (VCD) architecture is proposed to find question-critical scene temporally and disentangle the visual spurious correlations by attention-based front-door causal intervention module named Local-Global Causal Attention Module (LGCAM). To align the fine-grained interactions between linguistic semantics and spatial-temporal representations, we build an Interactive Visual-Linguistic Transformer (IVLT) that builds the multi-modal co-occurrence interactions between visual and linguistic content. Extensive experiments on four datasets demonstrate the superiority of CMQR for discovering visual causal structures and achieving robust question reasoning." @default.
- W4366388861 created "2023-04-21" @default.
- W4366388861 creator A5023363049 @default.
- W4366388861 creator A5042965510 @default.
- W4366388861 creator A5058965019 @default.
- W4366388861 date "2023-04-17" @default.
- W4366388861 modified "2023-09-28" @default.
- W4366388861 title "Causality-aware Visual Scene Discovery for Cross-Modal Question Reasoning" @default.
- W4366388861 doi "https://doi.org/10.48550/arxiv.2304.08083" @default.
- W4366388861 hasPublicationYear "2023" @default.
- W4366388861 type Work @default.
- W4366388861 citedByCount "0" @default.
- W4366388861 crossrefType "posted-content" @default.
- W4366388861 hasAuthorship W4366388861A5023363049 @default.
- W4366388861 hasAuthorship W4366388861A5042965510 @default.
- W4366388861 hasAuthorship W4366388861A5058965019 @default.
- W4366388861 hasBestOaLocation W43663888611 @default.
- W4366388861 hasConcept C105795698 @default.
- W4366388861 hasConcept C111472728 @default.
- W4366388861 hasConcept C115086926 @default.
- W4366388861 hasConcept C11671645 @default.
- W4366388861 hasConcept C119857082 @default.
- W4366388861 hasConcept C121332964 @default.
- W4366388861 hasConcept C138885662 @default.
- W4366388861 hasConcept C154945302 @default.
- W4366388861 hasConcept C15744967 @default.
- W4366388861 hasConcept C163504300 @default.
- W4366388861 hasConcept C166151441 @default.
- W4366388861 hasConcept C169760540 @default.
- W4366388861 hasConcept C169900460 @default.
- W4366388861 hasConcept C184337299 @default.
- W4366388861 hasConcept C185592680 @default.
- W4366388861 hasConcept C188027245 @default.
- W4366388861 hasConcept C188147891 @default.
- W4366388861 hasConcept C199360897 @default.
- W4366388861 hasConcept C204321447 @default.
- W4366388861 hasConcept C2777508537 @default.
- W4366388861 hasConcept C2779662365 @default.
- W4366388861 hasConcept C33923547 @default.
- W4366388861 hasConcept C41008148 @default.
- W4366388861 hasConcept C62520636 @default.
- W4366388861 hasConcept C64357122 @default.
- W4366388861 hasConcept C71139939 @default.
- W4366388861 hasConcept C97256817 @default.
- W4366388861 hasConceptScore W4366388861C105795698 @default.
- W4366388861 hasConceptScore W4366388861C111472728 @default.
- W4366388861 hasConceptScore W4366388861C115086926 @default.
- W4366388861 hasConceptScore W4366388861C11671645 @default.
- W4366388861 hasConceptScore W4366388861C119857082 @default.
- W4366388861 hasConceptScore W4366388861C121332964 @default.
- W4366388861 hasConceptScore W4366388861C138885662 @default.
- W4366388861 hasConceptScore W4366388861C154945302 @default.
- W4366388861 hasConceptScore W4366388861C15744967 @default.
- W4366388861 hasConceptScore W4366388861C163504300 @default.
- W4366388861 hasConceptScore W4366388861C166151441 @default.
- W4366388861 hasConceptScore W4366388861C169760540 @default.
- W4366388861 hasConceptScore W4366388861C169900460 @default.
- W4366388861 hasConceptScore W4366388861C184337299 @default.
- W4366388861 hasConceptScore W4366388861C185592680 @default.
- W4366388861 hasConceptScore W4366388861C188027245 @default.
- W4366388861 hasConceptScore W4366388861C188147891 @default.
- W4366388861 hasConceptScore W4366388861C199360897 @default.
- W4366388861 hasConceptScore W4366388861C204321447 @default.
- W4366388861 hasConceptScore W4366388861C2777508537 @default.
- W4366388861 hasConceptScore W4366388861C2779662365 @default.
- W4366388861 hasConceptScore W4366388861C33923547 @default.
- W4366388861 hasConceptScore W4366388861C41008148 @default.
- W4366388861 hasConceptScore W4366388861C62520636 @default.
- W4366388861 hasConceptScore W4366388861C64357122 @default.
- W4366388861 hasConceptScore W4366388861C71139939 @default.
- W4366388861 hasConceptScore W4366388861C97256817 @default.
- W4366388861 hasLocation W43663888611 @default.
- W4366388861 hasOpenAccess W4366388861 @default.
- W4366388861 hasPrimaryLocation W43663888611 @default.
- W4366388861 hasRelatedWork W1767411471 @default.
- W4366388861 hasRelatedWork W1893161884 @default.
- W4366388861 hasRelatedWork W1978959036 @default.
- W4366388861 hasRelatedWork W1991956303 @default.
- W4366388861 hasRelatedWork W2903145669 @default.
- W4366388861 hasRelatedWork W2997391570 @default.
- W4366388861 hasRelatedWork W3013663607 @default.
- W4366388861 hasRelatedWork W3088104152 @default.
- W4366388861 hasRelatedWork W4243804444 @default.
- W4366388861 hasRelatedWork W4287823982 @default.
- W4366388861 isParatext "false" @default.
- W4366388861 isRetracted "false" @default.
- W4366388861 workType "article" @default.