/Writing-editing-Network

Code for Paper Abstract Writing through Editing Mechanism

Primary LanguagePythonMIT LicenseMIT

Writing-editing Network: Paper Abstract Writing through Editing Mechanism

Paper Abstract Writing through Editing Mechanism

[Poster][Fake Handbook*] *Fake abstracts for the main conference (ACL 2018)

Accpeted by 56th Annual Meeting of the Association for Computational Linguistics (ACL 2018)

Table of Contents

Overview

Photo

Requirements

Environment:

  • Pytorch 0.4
  • Python 3.6 CAUTION!! Model will not be saved and loaded properly under Python 3.5

Data:

Quickstart

Preprocessing:

Put the acl_titles_and_abstracts.txt under the Writing-editing network folder. Randomly split the data into train, dev and test by runing split_data.py.

Training

Hyperparameter can be adjust in the Config class of main.py.

python main.py --cuda --mode 0

Validation

Compute score:

python main.py --cuda --mode 3

Predict abstract:

python main.py --cuda --mode 1

Citation

@InProceedings{P18-2042,
  author = 	"Wang, Qingyun
		and Zhou, Zhihao
		and Huang, Lifu
		and Whitehead, Spencer
		and Zhang, Boliang
		and Ji, Heng
		and Knight, Kevin",
  title = 	"Paper Abstract Writing through Editing Mechanism",
  booktitle = 	"Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers)",
  year = 	"2018",
  publisher = 	"Association for Computational Linguistics",
  pages = 	"260--265",
  location = 	"Melbourne, Australia",
  url = 	"http://aclweb.org/anthology/P18-2042"
}