/multicast

Pytorch implementation of "Empower Distantly Supervised Relation Extraction with Collaborative Adversarial Training"

Primary LanguagePython

MULTICAST

Code base for "Empower Distantly Supervised Relation Extraction with Collaborative Adversarial Training".

Env

Refer to repo OpenNRE, use pip to install required libraries → requirements.txt.

Run

  1. Run script download.py to download dataset nyt and embedding glove.
python download.py
  1. Run script train.sh to train and evaluate the DSRE models.
bash train.sh SEED

Cite

@inproceedings{chen2021empower,
  title={Empower Distantly Supervised Relation Extraction with Collaborative Adversarial Training},
  author={Chen, Tao and Shi, Haochen and Liu, Liyuan and Tang, Siliang and Shao, Jian and Chen, Zhigang and Zhuang, Yueting},
  booktitle={Proceedings of the AAAI Conference on Artificial Intelligence},
  volume={35},
  number={14},
  pages={12675--12682},
  year={2021}
}