MMCI uses multi-relational graph modeling and attention-based disentanglement of causal versus shortcut features, combined with backdoor adjustment, to reduce bias and improve generalization in multimodal sentiment analysis.
In Proceedings of the ACM Web Confer- ence 2022, 3562–3571
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Disentangling Bias by Modeling Intra- and Inter-modal Causal Attention for Multimodal Sentiment Analysis
MMCI uses multi-relational graph modeling and attention-based disentanglement of causal versus shortcut features, combined with backdoor adjustment, to reduce bias and improve generalization in multimodal sentiment analysis.