Semi-supervised Visual Feature Integration for Pre-trained Language Models
This work addresses the challenge of costly data alignment for multimodal language learning, offering an efficient solution for researchers and practitioners in natural language understanding.
The paper tackles the problem of expensive visual-textual alignment in multimodal language models by proposing a semi-supervised visual integration framework that does not require aligned images for every sentence, resulting in improvements on natural language inference and reading comprehension tasks over baseline models.
Integrating visual features has been proved useful for natural language understanding tasks. Nevertheless, in most existing multimodal language models, the alignment of visual and textual data is expensive. In this paper, we propose a novel semi-supervised visual integration framework for pre-trained language models. In the framework, the visual features are obtained through a visualization and fusion mechanism. The uniqueness includes: 1) the integration is conducted via a semi-supervised approach, which does not require aligned images for every sentences 2) the visual features are integrated as an external component and can be directly used by pre-trained language models. To verify the efficacy of the proposed framework, we conduct the experiments on both natural language inference and reading comprehension tasks. The results demonstrate that our mechanism brings improvement to two strong baseline models. Considering that our framework only requires an image database, and no not requires further alignments, it provides an efficient and feasible way for multimodal language learning.