TCR
收藏arXiv2025-09-30 收录
下载链接：
https://github.com/cogcomp/tcr
下载链接
链接失效反馈官方服务：
资源简介：
该数据集是为表格内容重建（TCR）任务而设计的，包含的表格内容不超过500个标记。在序列长度分布上，平均每个序列含有213.95个标记。与类似任务相比，这个数据集的规模相对较小，其具体任务是进行表格内容重建（TCR）。
This dataset is designed for the Table Content Reconstruction (TCR) task, and each table included in it contains no more than 500 tokens. In terms of sequence length distribution, the average number of tokens per sequence is 213.95. Compared with datasets for similar tasks, this dataset is relatively small in scale, with its specific task being Table Content Reconstruction (TCR).
