Increasing Visual Awareness in Multimodal Neural Machine Translation from an Information Theoretic Perspective
This work addresses a specific bottleneck in multimodal machine translation for researchers and practitioners, but it is incremental as it builds on existing methods to enhance visual awareness.
The paper tackled the problem of input degradation in multimodal machine translation, where models overlook visual information, by decomposing visual signals into source-specific and target-specific information and optimizing with mutual information, achieving superior results on two datasets.
Multimodal machine translation (MMT) aims to improve translation quality by equipping the source sentence with its corresponding image. Despite the promising performance, MMT models still suffer the problem of input degradation: models focus more on textual information while visual information is generally overlooked. In this paper, we endeavor to improve MMT performance by increasing visual awareness from an information theoretic perspective. In detail, we decompose the informative visual signals into two parts: source-specific information and target-specific information. We use mutual information to quantify them and propose two methods for objective optimization to better leverage visual signals. Experiments on two datasets demonstrate that our approach can effectively enhance the visual awareness of MMT model and achieve superior results against strong baselines.