AIM Data Set

该数据集是从数学领域自动定理证明器发现的证明中提取的示例（AIM循环）。数据集被分为训练集（60%）、验证集（10%）和测试集（30%）。规模涵盖20个数据集（8个地面数据集，12个非地面数据集），示例数量从251到34,101不等。任务的目的是学习符号重写，以应用于地面和非地面术语。

This dataset comprises examples extracted from proofs discovered by automated theorem provers in the field of mathematics (AIM cycle). It is split into three subsets: training set (60%), validation set (10%), and test set (30%). This dataset collection includes 20 sub-datasets, namely 8 ground datasets and 12 non-ground datasets, with the number of examples ranging from 251 to 34,101. The task objective is to learn symbolic rewriting for application to both ground and non-ground terms.