Many self-supervised learning methods are pre-trained on the well-curated ImageNet-1K dataset. In this work, given the excellent scalability of web data, we consider self-supervised pre-training on noisy web sourced image-text paired data. First, we conduct a benchmark study of representative self-supervised pre-training methods on large-scale web data in a like-for-like setting. We compare a range of methods, including single-modal ones that use masked training objectives and multi-modal ones that use image-text constrastive training. We observe that existing multi-modal methods do not outperform their single-modal counterparts on vision transfer learning tasks. We derive an information-theoretical view to explain these benchmark results, which provides insight into how to design a novel vision learner. Inspired by this insight, we present a new visual representation pre-training method, MUlti-modal Generator~(MUG), that learns from scalable web sourced image-text data. MUG achieves state-of-the-art transfer performance on a variety of tasks and demonstrates promising scaling properties. Pre-trained models and code will be made public upon acceptance.
@article{arxiv.2301.07088,
title = {Vision Learners Meet Web Image-Text Pairs},
author = {Bingchen Zhao and Quan Cui and Hao Wu and Osamu Yoshie and Cheng Yang and Oisin Mac Aodha},
journal= {arXiv preprint arXiv:2301.07088},
year = {2024}
}