Deep Graph Convolutional Encoders for Structured Data to Text Generation

Diego Marcheggiani, Laura Perez-Beltrachini
10/23/2018
cs.CL

Official code: diegma/graph-2-text (152 stars, MIT, last commit 2019-07-12).

Abstract

Most previous work on neural text generation from graph-structured data relies on standard sequence-to-sequence methods. These approaches linearise the input graph to be fed to a recurrent neural network. In this paper, we propose an alternative encoder based on graph convolutional networks that directly exploits the input structure. We report results on two graph-to-sequence datasets that empirically show the benefits of explicitly encoding the input graph structure.

View on arXivView PDF

Code Implementations(1)

diegma/graph-2-textOfficial100%
MIT

Graph to sequence implemented in Pytorch combining Graph convolutional networks and opennmt-py

15228Perl, RubySep 25, 20187 years agoMIT

Similar Papers

Graph-to-Sequence Learning using Gated Graph Neural Networks

88% match1 repo1222018

Structural Neural Encoders for AMR-to-text Generation

86% match1 repo232019

Graph2Seq: Graph to Sequence Learning with Attention-based Neural Networks

85% match1 repo2402018

Structured Neural Summarization

83% match1 repo722018

Tensor Graph Convolutional Networks for Text Classification

81% match1 repo292020

Harnessing Explanations: LLM-to-LM Interpreter for Enhanced Text-Attributed Graph Representation Learning

81% match2 repos2722023

Cite this paper

@article{marcheggiani2018deep,
  title  = {Deep Graph Convolutional Encoders for Structured Data to Text Generation},
  author = {Diego Marcheggiani and Laura Perez-Beltrachini},
  year   = {2018},
  eprint = {1810.09995},
  archivePrefix = {arXiv},
  url    = {https://arxiv.org/abs/1810.09995}
}

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/1810.09995.svg)](https://codeofpaper.com/paper/1810.09995)

Discussion