THU-KEG/MMLongBench-Write

该数据集包含四个字段：id、question、L和images。id是唯一标识符，question是字符串类型的问题，L是整数类型的某个长度，images是图像序列。数据集有一个训练集（train），共有120个示例，总大小为46727796字节。具体的应用场景或数据集内容未在README中描述。The dataset includes four fields: id, question, L, and images. The id is a unique identifier, question is a string type question, L is an integer type length, and images is an image sequence. The dataset has a training set (train) with a total of 120 examples and a total size of 46727796 bytes. The specific application scenario or content of the dataset is not described in the README.