VLMo: Unified Vision-Language Pre-Training with Mixture-of-Modality-Experts
This addresses the need for flexible and efficient vision-language models for AI applications, though it is incremental as it builds on existing Transformer and pre-training methods.
The paper tackles the problem of vision-language pre-training by introducing VLMo, a unified model that uses a Mixture-of-Modality-Experts Transformer to jointly learn dual and fusion encoders, achieving state-of-the-art results on tasks like VQA, NLVR2, and image-text retrieval.
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 self-attention 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.