VLMo: Unified Vision-Language Pre-Training with Mixture-of-Modality-Experts

We present a unified Vision-Language pretrained Model (VLMO) that jointly learns a dual encoder and a fusion encoder with a modular Transformer network. Specifically, we introduce Mixture-of-Modality-Experts (MOME) Transformer, where each block contains a pool of modality-specific experts and a shared selfattention layer. Because of the modeling flexibility of MOME, pretrained VLMO can be fine-tuned as a fusion encoder for vision-language classification tasks, or used as a dual encoder for efficient image-text retrieval. Moreover, we propose a stagewise pre-training strategy, which effectively leverages large-scale image-only and text-only data besides image-text pairs. Experimental results show that VLMO achieves state-of-the-art results on various vision-language tasks, including VQA, NLVR2 and image-text retrieval. The code and pretrained models are available at https://aka.ms/vlmo . * Equal contribution. † Contact person.