Global Encoding for Abstractive Summarization

Junyang Lin, Xu Sun, Shuming Ma, Qi Su
5/10/2018
cs.CLcs.AIcs.LG

Official code: lancopku/Global-Encoding (273 stars, MIT, last commit 2019-05-30).

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.

View on arXivView PDF

Code Implementations(1)

lancopku/Global-EncodingOfficial100%
MIT

Global Encoding for Abstractive Summarization (ACL 2018)

27363HTML, PythonMay 10, 20187 years agoMIT

Similar Papers

Selective Encoding for Abstractive Sentence Summarization

88% match1 repo372017

Neural Abstractive Text Summarization with Sequence-to-Sequence Models

86% match1 repo1592018

Improving Sequence-to-Sequence Models for Abstractive Text Summarization Using Meta Heuristic Approaches

85% match2024

Neural Sequence-to-Sequence Modeling with Attention by Leveraging Deep Learning Architectures for Enhanced Contextual Understanding in Abstractive Text Summarization

83% match2024

Get To The Point: Summarization with Pointer-Generator Networks

83% match1 repo2.2k2017

MeanSum: A Neural Model for Unsupervised Multi-document Abstractive Summarization

82% match1 repo1142018

Cite this paper

@article{lin2018global,
  title  = {Global Encoding for Abstractive Summarization},
  author = {Junyang Lin and Xu Sun and Shuming Ma and Qi Su},
  year   = {2018},
  eprint = {1805.03989},
  archivePrefix = {arXiv},
  url    = {https://arxiv.org/abs/1805.03989}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/1805.03989.svg)](https://codeofpaper.com/paper/1805.03989)

Discussion