MINT-1T: Scaling Open-Source Multimodal Data by 10x: A Multimodal Dataset with One Trillion Tokens

Published: 26 Sept 2024, Last Modified: 13 Nov 2024NeurIPS 2024 Track Datasets and Benchmarks PosterEveryoneRevisionsBibTeXCC BY 4.0
Keywords: multimodal data, open-source, vision-language models
TL;DR: MINT-1T is the first open-source trillion token multimodal interleaved dataset from diverse sources.
Abstract: Multimodal interleaved datasets featuring free-form interleaved sequences of images and text are crucial for training frontier large multimodal models (LMMs). Despite the rapid progression of open-source LMMs, there remains a pronounced scarcity of large-scale, open-source multimodal interleaved datasets. In response, we introduce MINT-1T, the most extensive and diverse open-source Multimodal INTerleaved dataset to date. MINT-1T comprises of one trillion text tokens and 3.4 billion images, a 10x scale-up from existing open-source datasets. Additionally, we include previously untapped sources such as PDFs and ArXiv papers. As scaling multimodal interleaved datasets requires substantial engineering effort, sharing the data curation process and releasing the dataset greatly benefits the community. Our experiments show that LMMs trained on MINT-1T rival the performance of models trained on the previous leading dataset, OBELICS. We release our data at https://github.com/mlfoundations/MINT-1T.
Submission Number: 2146
Loading