Vision-and-language pre-training has achieved impressive success in learning multimodal representations between vision and language. To generalize this success to non-English languages, we introduce UC2, the first machine translation-augmented framework for cross-lingual cross-modal representation learning. To tackle the scarcity problem of multilingual captions for image datasets, we first augment existing English-only datasets with other languages via machine translation (MT). Then we extend the standard Masked Language Modeling and Image-Text Matching training objectives to multilingual setting, where alignment between different languages is captured through shared visual context (i.e, using image as pivot). To facilitate the learning of a joint embedding space of images and all languages of interest, we further propose two novel pre-training tasks, namely Masked Region-to-Token Modeling (MRTM) and Visual Translation Language Modeling (VTLM), leveraging MT-enhanced translated data. Evaluation on multilingual image-text retrieval and multilingual visual question answering benchmarks demonstrates that our proposed framework achieves new state-of-the-art on diverse non-English benchmarks while maintaining comparable performance to monolingual pre-trained models on English tasks.

UC2是第一个基于机器翻译增强的框架，用于跨语言跨模态表示学习。我们扩充了现有的只有英语的数据集，通过机器翻译引入了其他语言的图像标题，然后将标准的Masked Language Modeling 和 Image-Text Matching训练目标扩展到多语言环境，通过共享视觉上下文（即使用图像作为枢纽）来捕获不同语言之间的对齐。最终我们提出了两个新的预训练任务，Masked Region-to-Token Modeling（MRTM）和Visual Translation Language Modeling（VTLM），以加快图像和所有感兴趣语言的联合嵌入空间的学习。在多语言图像文本检索和多语言视觉问答基准上的评估表明，我们的提议框架在各种非英语基准上实现了新的最先进状态，并在英语任务上保持与单语预训练模型相当的性能。

UC2: 通用跨语言跨模态视觉语言预训练