Like most natural language understanding and generation tasks, state-of-the-art models for summarization are transformer-based sequence-to-sequence architectures that are pretrained on large corpora. While most existing models focused on English, Arabic remained understudied. In this paper we propose AraBART, the first Arabic model in which the encoder and the decoder are pretrained end-to-end, based on BART. We show that AraBART achieves the best performance on multiple abstractive summarization datasets, outperforming strong baselines including a pretrained Arabic BERT-based model and multilingual mBART and mT5 models.
@article{arxiv.2203.10945,
title = {AraBART: a Pretrained Arabic Sequence-to-Sequence Model for Abstractive Summarization},
author = {Moussa Kamal Eddine and Nadi Tomeh and Nizar Habash and Joseph Le Roux and Michalis Vazirgiannis},
journal= {arXiv preprint arXiv:2203.10945},
year = {2022}
}