Research Area:  Machine Learning

Abstract:

BERT, a pre-trained Transformer model, has achieved ground-breaking performance on multiple NLP tasks. In this paper, we describe BERTSUM, a simple variant of BERT, for extractive summarization. Our system is the state of the art on the CNN/Dailymail dataset, outperforming the previous best-performed system by 1.65 on ROUGE-L.

Keywords:  

Author(s) Name:  Yang Liu

Publisher name:  arXiv:1903.10318

DOI:  10.48550/arXiv.1903.10318

Volume Information:  

Paper Link:   https://arxiv.org/abs/1903.10318