Existing models for extractive summarization are usually trained from scratch with a cross-entropy loss, which does not explicitly capture the global context at the document level. In this paper, we aim to improve this task by introducing three auxiliary pre-training tasks that learn to capture the document-level context in a self-supervised fashion. Experiments on the widely-used CNN/DM dataset validate the effectiveness of the proposed auxiliary tasks. Furthermore, we show that after pre-training, a clean model with simple building blocks is able to outperform previous state-of-the-art that are carefully designed.
@article{arxiv.1906.04466,
title = {Self-Supervised Learning for Contextualized Extractive Summarization},
author = {Hong Wang and Xin Wang and Wenhan Xiong and Mo Yu and Xiaoxiao Guo and Shiyu Chang and William Yang Wang},
journal= {arXiv preprint arXiv:1906.04466},
year = {2019}
}