THU-KEG/MM-Math-Align

MM-Math-Align数据集是基于中学几何考试中实际使用的几何问题构建的。每个样本包含原始几何图表（original_image）、一个Python脚本的图像（positive_image）近似重建图表、一个描述正图像的标题（positive_caption）、基于正图像生成的10个负Python脚本图像（negative_idx_image）以及相应的10个负标题（negative_idx_caption）。该数据集总共包含4021个样本。

We release MM-Math-Align, a dataset built upon MM-Math, which is derived from actual geometry questions used in middle school exams. Each sample contains the original geometric diagram (original_image), a Python scripts image (positive_image) that approximately reconstructs the diagram, a caption (positive_caption) describing the positive image, 10 negative Python script images (negative_idx_image) generated based on the positive image, and 10 corresponding negative captions (negative_idx_caption). The dataset consists of a total of 4,021 samples.