octopus

<p align="center"> <br> <img src="https://github.com/UBC-NLP/octopus/raw/main/images/octopus.jpg" width="70%" height="70%"/> <br> <p> <img src="https://github.com/UBC-NLP/octopus/raw/main/images/octopus_tasks.png" alt="OCTOPUS" width="55%" height="55%" align="right"/>

Octopus is a neural machine generation toolkit for Arabic Natural Lnagauge Generation (NLG) that described in our ArabiNLP 2023 paper: OCTOPUS: A Multitask Model and Toolkit for Arabic Natural Language Generation.

Octopus designed for eight machine generation tasks, encompassing diacritization, grammatical error correction, news headlines generation, paraphrasing, question answering, question generation, and transliteration. This comprehensive package includes a Python library along with associated command-line scripts.

Github: https://github.com/UBC-NLP/octopus

Demo: https://demos.dlnlp.ai/octopus

Paper: https://arxiv.org/abs/2206.03933

License

octopus(-py) is Apache-2.0 licensed. The license applies to the pre-trained models as well.

Citation

If you use TURJUMAN toolkit or the pre-trained models for your scientific publication, or if you find the resources in this repository useful, please cite our paper as follows (to be updated):

 @misc{elmadany2023octopus,
      title={Octopus: A Multitask Model and Toolkit for Arabic Natural Language Generation}, 
      author={AbdelRahim Elmadany and El Moatez Billah Nagoudi and Muhammad Abdul-Mageed},
      year={2023},
      eprint={2310.16127},
      archivePrefix={arXiv},
      primaryClass={cs.CL}
}