SED_Aug - ACL2025

The official code repository of SED-Aug model from ACL 2025 paper "Explicit and Implicit Data Augmentation for Social Event Detection"

Installation

pip install -r requirements.txt

For more requirements, please refer to requirements.txt.

Data preparation

For Kawarith6 dataset, please refer to paper "Kawarith: an Arabic Twitter Corpus for Crisis Events" For Twitter2012 dataset, please refer to paper "Building a large-scale corpus for evaluating event detection on twitter" For Twitter2018 dataset, please refer to paper "A french corpus for event detection on twitter"

Model Training

The code is in the "code" folder. For model training, the commandline is:

bash run.sh [GPU id]

For instance:

bash run.sh 0

Acknowledgement

If you got a chance to use our code, you could consider to cite our paper with the following information:

@inproceedings{Ma2025Explicit,
  author       = {Congbo Ma and
                  Yuxia Wang and
                  Jia Wu and
                  Jian Yang and
                  Jing Du and
                  Zitai Qiu and
                  Qing Li and
                  Hu Wang and
                  Preslav Nakov},
  title        = {Explicit and Implicit Data Augmentation for Social Event Detection},
  booktitle    = {Proceedings of the 63rd Annual Meeting of the Association for Computational
                  Linguistics (Volume 1: Long Papers), {ACL} 2025, Vienna, Austria,
                  July 27 - August 1, 2025},
  pages        = {8402--8415},
  publisher    = {Association for Computational Linguistics},
  year         = {2025},
  url          = {https://aclanthology.org/2025.acl-long.412/}
}