A Comparison of Neural Models for Word Ordering
Computation and Language
2017-08-08 v1
Abstract
We compare several language models for the word-ordering task and propose a new bag-to-sequence neural model based on attention-based sequence-to-sequence models. We evaluate the model on a large German WMT data set where it significantly outperforms existing models. We also describe a novel search strategy for LM-based word ordering and report results on the English Penn Treebank. Our best model setup outperforms prior work both in terms of speed and quality.
Cite
@article{arxiv.1708.01809,
title = {A Comparison of Neural Models for Word Ordering},
author = {Eva Hasler and Felix Stahlberg and Marcus Tomalin and Adri`a de Gispert and Bill Byrne},
journal= {arXiv preprint arXiv:1708.01809},
year = {2017}
}
Comments
Accepted for publication at INLG 2017