MT

该数据集包含了参与者们通过口头协作，使用彼此的地图来描述路线的情景。其性能取决于训练集、开发集和测试集的划分方式。该数据集规模较小，包含约27,000个话语。其任务是进行对话行为分类。

This dataset encompasses scenarios where participants collaboratively described routes by utilizing each other’s maps through verbal cooperation. Its performance relies on the partitioning approach of the training, development, and test sets. This is a small-scale dataset containing approximately 27,000 utterances. The task of this dataset is dialogue act classification.