French Abstractive Summarization Dataset.
The dataset introduced in BARThez: a Skilled Pretrained French Sequence-to-Sequence Model.
Our project repo: https://github.com/moussaKam/BARThez
You can find the dataset and the code to reproduce it at: https://github.com/Tixierae/OrangeSum
If you use OrangeSum please cite us at:
@article{eddine2020barthez,
title={BARThez: a Skilled Pretrained French Sequence-to-Sequence Model},
author={Eddine, Moussa Kamal and Tixier, Antoine J-P and Vazirgiannis, Michalis},
journal={arXiv preprint arXiv:2010.12321},
year={2020}
}