/gtt

Template Filling with Generative Transformers

Primary LanguagePythonMIT LicenseMIT

Template Filling with Generative Transformers (NAACL 2021 short)

paper link

Dependencies

  • Python 3.6.10
  • Transformers: transformers 2.4.1 installed from source.
  • Pytorch-Lightning==0.7.1
  • Pytorch==1.4.0
  • seqeval

Dataset

  • ./data/muc, refer to ./data/muc/README.md for details

Eval

  • eval on preds_gtt.out python eval.py --pred_file model_gtt/preds_gtt.out

GTT model

Citation

If you use our materials, please cite

@inproceedings{du2021gtt,
  title={Template Filling with Generative Transformers},
  author={Du, Xinya and Rush, Alexander M and Cardie, Claire},
  booktitle = "Proceedings of the 2021 Conference of the North {A}merican Chapter of the Association for Computational Linguistics: Human Language Technologies",
  year={2021}
}