LAION-5B: An open large-scale dataset for training next generation image-text models

TL;DR

LAION-5B is a large, openly available dataset of 5.85 billion image-text pairs designed to facilitate research and development of next-generation large-scale multi-modal models like CLIP and DALL-E.

Contribution

The paper introduces LAION-5B, the largest open dataset of its kind, enabling broader access and experimentation with large-scale image-text models.

Findings

Successful replication of models like CLIP and Stable Diffusion using LAION-5B.

Demonstrated fine-tuning and transfer learning capabilities on the dataset.

Provided tools for dataset exploration and content detection.

Abstract

Groundbreaking language-vision architectures like CLIP and DALL-E proved the utility of training on large amounts of noisy image-text data, without relying on expensive accurate labels used in standard vision unimodal supervised learning. The resulting models showed capabilities of strong text-guided image generation and transfer to downstream tasks, while performing remarkably at zero-shot classification with noteworthy out-of-distribution robustness. Since then, large-scale language-vision models like ALIGN, BASIC, GLIDE, Flamingo and Imagen made further improvements. Studying the training and capabilities of such models requires datasets containing billions of image-text pairs. Until now, no datasets of this size have been made openly available for the broader research community. To address this problem and democratize research on large-scale multi-modal models, we present LAION-5B - a dataset consisting of 5.85 billion CLIP-filtered image-text pairs, of which 2.32B contain English language. We show successful replication and fine-tuning of foundational models like CLIP, GLIDE and Stable Diffusion using the dataset, and discuss further experiments enabled with an openly available dataset of this scale. Additionally we provide several nearest neighbor indices, an improved web-interface for dataset exploration and subset generation, and detection scores for watermark, NSFW, and toxic content detection. Announcement page https://laion.ai/laion-5b-a-new-era-of-open-large-scale-multi-modal-datasets/