Research Area:  Machine Learning
Neural extractive summarization models usually employ a hierarchical encoder for document encoding and they are trained using sentence-level labels, which are created heuristically using rule-based methods. Training the hierarchical encoder with these emph{inaccurate} labels is challenging. Inspired by the recent work on pre-training transformer sentence encoders cite{devlin:2018:arxiv}, we propose {sc Hibert} (as shorthand for {f HI}erachical {f B}idirectional {f E}ncoder {f R}epresentations from {f T}ransformers) for document encoding and a method to pre-train it using unlabeled data. We apply the pre-trained {sc Hibert} to our summarization model and it outperforms its randomly initialized counterpart by 1.25 ROUGE on the CNN/Dailymail dataset and by 2.0 ROUGE on a version of New York Times dataset. We also achieve the state-of-the-art performance on these two datasets.
Keywords:  
Author(s) Name:  Xingxing Zhang, Furu Wei, Ming Zhou
Journal name:  Computer Science
Conferrence name:  
Publisher name:  arXiv:1905.06566
DOI:  10.48550/arXiv.1905.06566
Volume Information:  
Paper Link:   https://arxiv.org/abs/1905.06566