CLTS+: A New Chinese Long Text Summarization Dataset with Abstractive Summaries

TL;DR

This paper introduces CLTS+, a large Chinese long text summarization dataset with highly abstractive summaries, aiming to enhance the creative ability of summarization models and providing a new benchmark for Chinese NLP research.

Contribution

The creation of CLTS+, the first Chinese long text summarization dataset with high abstractiveness, along with an intrinsic co-occurrence metric for evaluation and baseline experiments demonstrating its utility.

Findings

CLTS+ contains over 180K article-summary pairs.

The dataset exhibits high levels of abstractiveness and difficulty.

Baseline models trained on CLTS+ show improved creative summarization abilities.

Abstract

The abstractive methods lack of creative ability is particularly a problem in automatic text summarization. The summaries generated by models are mostly extracted from the source articles. One of the main causes for this problem is the lack of dataset with abstractiveness, especially for Chinese. In order to solve this problem, we paraphrase the reference summaries in CLTS, the Chinese Long Text Summarization dataset, correct errors of factual inconsistencies, and propose the first Chinese Long Text Summarization dataset with a high level of abstractiveness, CLTS+, which contains more than 180K article-summary pairs and is available online. Additionally, we introduce an intrinsic metric based on co-occurrence words to evaluate the dataset we constructed. We analyze the extraction strategies used in CLTS+ summaries against other datasets to quantify the abstractiveness and difficulty of our new data and train several baselines on CLTS+ to verify the utility of it for improving the creative ability of models.