ko-vlm/K-SEED
收藏官方服务：
资源简介：
该数据集包含图像和对话两种类型的数据。对话数据由内容字符串和角色字符串组成。数据集包含一个测试集，共有2971个示例，总大小为512,671,276.625字节。数据集的下载大小为505,008,645字节。
The dataset includes two types of data: images and conversations. The conversation data consists of content strings and role strings. The dataset contains a test split with 2971 examples, totaling 512,671,276.625 bytes in size. The download size of the dataset is 505,008,645 bytes.
提供机构：
ko-vlm
