Ti-SUM

Automatic text summarization is a key task in natural language processing. High-quality datasets can effectively promote the research progress of summarization. Recent research is closer to generate abstractive summarizations by using the deep learning methods. However, there is a lack of high-quality and large-scale summarization datasets available to the public. Besides, it is difficult to construct this kind of dataset manually. The Tibetan text summarization task is still in its infancy due to the lack of public datasets. In order to promote the development of Tibetan informatization. we artificially constructed a small dataset of Tibetan text summarization in this paper, which is composed of 1,000 real Tibetan news articles, each with a short summary. In addition, we have also constructed more than 3,500 article keywords for each news article as a supplement to text summarization tasks.