OmniCorpus: A Unified Multimodal Corpus of 10 Billion-Level Images Interleaved with Text

TL;DR

OmniCorpus is a massive 10 billion-scale image-text interleaved dataset that enhances multimodal model training by providing diverse, high-quality, and flexible data sources, supporting advanced research in multimodal learning.

Contribution

The paper introduces OmniCorpus, the largest and most diverse image-text interleaved dataset to date, enabling improved multimodal large language models.

Findings

Dataset contains 8.6 billion images and 1,696 billion text tokens.

OmniCorpus is 15 times larger than comparable datasets.

The dataset is versatile, supporting various data formats and sources.

Abstract

Image-text interleaved data, consisting of multiple images and texts arranged in a natural document format, aligns with the presentation paradigm of internet data and closely resembles human reading habits. Recent studies have shown that such data aids multimodal in-context learning and maintains the capabilities of large language models during multimodal fine-tuning. However, the limited scale and diversity of current image-text interleaved data restrict the development of multimodal large language models. In this paper, we introduce OmniCorpus, a 10 billion-scale image-text interleaved dataset. Using an efficient data engine, we filter and extract large-scale high-quality documents, which contain 8.6 billion images and 1,696 billion text tokens. Compared to counterparts (e.g., MMC4, OBELICS), our dataset 1) has 15 times larger scales while maintaining good data quality; 2) features more diverse sources, including both English and non-English websites as well as video-centric websites; 3) is more flexible, easily degradable from an image-text interleaved format to pure text corpus and image-text pairs. Through comprehensive analysis and experiments, we validate the quality, usability, and effectiveness of the proposed dataset. We hope this could provide a solid data foundation for future multimodal model research. Code and data are released at https://github.com/OpenGVLab/OmniCorpus.