Bamboo

该数据集是一个多任务长文本基准测试，包含了来自五个不同长文本理解任务的10个数据集。这些任务涵盖了诸如问答和虚构检测等领域，旨在对长文本建模进行深入研究和评估。

This dataset is a multi-task long-text benchmark that encompasses 10 datasets across five distinct long-text understanding tasks. These tasks cover domains such as question answering and fictional content detection, and the benchmark is designed to support in-depth research and evaluation of long-text modeling.