The increased interest in diffusion models has opened up opportunities for
advancements in generative text modeling. These models can produce impressive
images when given a well-crafted prompt, but creating a powerful or meaningful
prompt can be hit-or-miss. To address this, we have created a large-scale
dataset that is derived and synthesized from real prompts and indexed with
popular image-text datasets such as MS-COCO and Flickr. We have also
implemented stages that gradually reduce context and increase complexity, which
will further enhance the output due to the complex annotations created. The
dataset, called MTTN, includes over 2.4 million sentences divided into 5
stages, resulting in a total of over 12 million pairs, and a vocabulary of over
300,000 unique words, providing ample variation. The original 2.4 million pairs
are designed to reflect the way language is used on the internet globally,
making the dataset more robust for any model trained on it.

为了提高生成文本模型的效果，本研究创建了一个衍生且合成自真实提示的、与 Microsoft-COCO 和 Flickr 等流行图像 - 文本数据集索引的大规模数据集 MTTN，其包括超过 240 万个句子，分成 5 个阶段，共计超过 1200 万个对，并且具有超过 300,000 个独有词汇，目的是反映全球互联网语言的使用方式，从而使更强大的生成文本模型更能够利用。