Provable Dynamic Fusion for Low-Quality Multimodal Data
This work addresses the challenge of handling low-quality multimodal data in fusion tasks, providing theoretical justification and practical improvements for applications relying on such data.
The paper tackles the problem of robust multimodal fusion for low-quality data by proposing a Quality-aware Multimodal Fusion (QMF) framework, which improves classification accuracy and model robustness as demonstrated through extensive experiments on multiple benchmarks.
The inherent challenge of multimodal fusion is to precisely capture the cross-modal correlation and flexibly conduct cross-modal interaction. To fully release the value of each modality and mitigate the influence of low-quality multimodal data, dynamic multimodal fusion emerges as a promising learning paradigm. Despite its widespread use, theoretical justifications in this field are still notably lacking. Can we design a provably robust multimodal fusion method? This paper provides theoretical understandings to answer this question under a most popular multimodal fusion framework from the generalization perspective. We proceed to reveal that several uncertainty estimation solutions are naturally available to achieve robust multimodal fusion. Then a novel multimodal fusion framework termed Quality-aware Multimodal Fusion (QMF) is proposed, which can improve the performance in terms of classification accuracy and model robustness. Extensive experimental results on multiple benchmarks can support our findings.