Sigmoid Loss for Language Image Pre-Training

Abstract

We propose a simple pairwise sigmoid loss for image-text pre-training. Unlike standard contrastive learning with softmax normalization, the sigmoid loss operates solely on image-text pairs and does not require a global view of the pairwise similarities for normalization. The sigmoid loss simultaneously allows further scaling up the batch size, while also performing better at smaller batch sizes. With only four TPUv4 chips, we can train a Base CLIP model at 4k batch size and a Large LiT model at 20k batch size, the latter achieves 84.5% ImageNet zero-shot accuracy in two days. This disentanglement of the batch size from the loss further allows us to study the impact of examples vs pairs and negative to positive ratio. Finally, we push the batch size to the extreme, up to one million, and find that the benefits of growing batch size quickly diminish, with a more reasonable batch size of 32k being sufficient. We hope our research motivates further explorations in improving the quality and efficiency of language-image pre-training.

Full paper on arXiv · Code

Details

arXiv identifier

2303.15343

Published

2023-03-27

Authors

Xiaohua Zhai, Basil Mustafa, Alexander Kolesnikov, Lucas Beyer

Models That Cite This Paper

• Described bysiglip2-base-patch16-256
• Described bysiglip2-giant-opt-patch16-384
• Described bysiglip-base-patch16-224
• Described bysiglip2-base-patch16-224
• Described bysiglip-so400m-patch14-384
• Described bymedgemma-4b-it
• Described bysiglip2-so400m-patch14-384
• Described bysiglip2-base-patch16-naflex