WMT32k

该数据集是由一系列年度机器翻译研讨会（WMT）提供的，包含了多种语言对的资料集。它被用于对Transformer基础模型和大型模型进行全面的训练。该数据集主要针对的自然语言处理任务为翻译。

This dataset is derived from a series of annual Workshop on Machine Translation (WMT) events, and comprises parallel corpora across multiple language pairs. It is utilized for comprehensive training of both base and large-scale Transformer models. The primary natural language processing (NLP) task targeted by this dataset is machine translation.