Source code for ACL 2023 paper: An AMR-based Link Prediction Approach for Document-level Event Argument Extraction.
pip install git+https://github.com/fastnlp/fastNLP@dev0.8.0
pip install git+https://github.com/fastnlp/fitlog
conda install pytorch==1.9.0 torchvision==0.10.0 torchaudio==0.9.0 cudatoolkit=10.2 -c pytorch
pip install transformers==4.22.2
pip install dgl-cu102 dglgo -f https://data.dgl.ai/wheels/repo.html
We provide the preprocessed data here, which can be downloaded and used directly.
If you need to preprocess data from text, please refer to data_processing.
Please execute the command python src_x/train.py
for RAMS or WikiEvents, separately. To make adjustments to hyperparameters, kindly refer to src_x/parse.py
and implement any necessary modifications.
You can evaluate the trained model by running the following commands:
bash evaluate_rams.sh
bash evaluate_wikievents.sh
If you find our work useful, please cite our paper:
@inproceedings{DBLP:conf/acl/0004GHZQZ23,
author = {Yuqing Yang and
Qipeng Guo and
Xiangkun Hu and
Yue Zhang and
Xipeng Qiu and
Zheng Zhang},
editor = {Anna Rogers and
Jordan L. Boyd{-}Graber and
Naoaki Okazaki},
title = {An AMR-based Link Prediction Approach for Document-level Event Argument
Extraction},
booktitle = {Proceedings of the 61st Annual Meeting of the Association for Computational
Linguistics (Volume 1: Long Papers), {ACL} 2023, Toronto, Canada,
July 9-14, 2023},
pages = {12876--12889},
publisher = {Association for Computational Linguistics},
year = {2023},
url = {https://doi.org/10.18653/v1/2023.acl-long.720},
doi = {10.18653/v1/2023.acl-long.720},
timestamp = {Thu, 10 Aug 2023 12:35:57 +0200},
biburl = {https://dblp.org/rec/conf/acl/0004GHZQZ23.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}