English

Global Encoding for Abstractive Summarization

Computation and Language 2018-06-14 v2 Artificial Intelligence Machine Learning

Abstract

In neural abstractive summarization, the conventional sequence-to-sequence (seq2seq) model often suffers from repetition and semantic irrelevance. To tackle the problem, we propose a global encoding framework, which controls the information flow from the encoder to the decoder based on the global information of the source context. It consists of a convolutional gated unit to perform global encoding to improve the representations of the source-side information. Evaluations on the LCSTS and the English Gigaword both demonstrate that our model outperforms the baseline models, and the analysis shows that our model is capable of reducing repetition.

Keywords

Cite

@article{arxiv.1805.03989,
  title  = {Global Encoding for Abstractive Summarization},
  author = {Junyang Lin and Xu Sun and Shuming Ma and Qi Su},
  journal= {arXiv preprint arXiv:1805.03989},
  year   = {2018}
}

Comments

Accepted by ACL 2018