Code related to our AACL-IJCNLP 2022 submission
- Data : All the datasets (originals, generated, structure, etc.)
- Models : Our implementations of the ConvS2S and Transformer architectures as well as the Colab notebooks to evaluate and infer
If you use our work, please cite the following paper:
Rose Hirigoyen, Amal Zouaq, and Samuel Reyd. 2022. A Copy Mechanism for Handling Knowledge Base Elements in SPARQL Neural Machine Translation. In Findings of the Association for Computational Linguistics: AACL-IJCNLP 2022, pages 226–236, Online only. Association for Computational Linguistics.
Check our presentation at: http://www.labowest.ca/conferences/AACL2022/Hirigoyen_et_al_AACL-IJCNLP2022.mp4