UNIMO: Towards Unified-Modal Understanding and Generation via Cross-Modal Contrastive Learning
This work addresses the limitation of existing pre-training methods that struggle to adapt between single-modal and multi-modal tasks, offering a more versatile solution for researchers and practitioners in AI.
The paper introduces UNIMO, a unified-modal pre-training architecture designed to handle both single-modal and multi-modal understanding and generation tasks. It leverages large-scale text and image collections, along with cross-modal contrastive learning, to align textual and visual information into a unified semantic space, resulting in significant performance improvements across various downstream tasks.
Existed pre-training methods either focus on single-modal tasks or multi-modal tasks, and cannot effectively adapt to each other. They can only utilize single-modal data (i.e. text or image) or limited multi-modal data (i.e. image-text pairs). In this work, we propose a unified-modal pre-training architecture, namely UNIMO, which can effectively adapt to both single-modal and multi-modal understanding and generation tasks. Large scale of free text corpus and image collections can be utilized to improve the capability of visual and textual understanding, and cross-modal contrastive learning (CMCL) is leveraged to align the textual and visual information into a unified semantic space over a corpus of image-text pairs. As the non-paired single-modal data is very rich, our model can utilize much larger scale of data to learn more generalizable representations. Moreover, the textual knowledge and visual knowledge can enhance each other in the unified semantic space. The experimental results show that UNIMO significantly improves the performance of several single-modal and multi-modal downstream tasks. Our code and pre-trained models are public at the UNIMO project page https://unimo-ptm.github.io/