anthracite-org/pixmo-cap-images

PixMo-Cap是一个包含详细长文本描述的数据集，主要用于预训练和微调视觉-语言模型。数据集中的描述是通过记录注释者对图像的60-90秒语音描述，并使用Claude大型语言模型将音频转录转换为长文本描述。数据集还包括音频转录。此外，数据集是PixMo数据集集合的一部分，并用于训练Molmo系列模型。数据集格式包括图像、图像URL、描述和转录。

PixMo-Cap is a dataset of very long (roughly 200 words on average), detailed captions for image-to-text tasks. It is used for pre-training and fine-tuning vision-language models. The dataset was created by recording annotators speaking about an image for 60-90 seconds and then using the Claude large language model to turn the audio transcripts into long captions. The audio transcripts are also included. PixMo-Cap is part of the PixMo dataset collection and was used to train the Molmo family of models. Unlike the original release, images are included in the dataset itself.