Affective Event Generation

Werkzeug

Yarik Menchaca Resendiz, Roman Klinger

This repository contains code and data for the fine-tuning of T5 / Bart models to generate emotional event descriptions, as presented in the paper. There are three model configurations: condition on emotions (E), condition on emotions and appraisal (EA), and condition on appraisals (A).

Yarik Menchaca Resendiz and Roman Klinger. Affective natural language generation of event descriptions through fine-grained appraisal conditions. In Proceedings of the 16th International Conference on Natural Language Generation, Prague, Czech Republic, September 2023. Association for Computational Linguistics. accepted.

Citation:

@inproceedings{MenchacaResendiz2023,
  title = {Affective Natural Language Generation of Event
                  Descriptions through Fine-grained Appraisal
                  Conditions},
  author = {Menchaca Resendiz, Yarik and Klinger, Roman},
  booktitle = {Proceedings of the 16th International Conference on
                  Natural Language Generation},
  month = sep,
  year = {2023},
  address = {Prague, Czech Republic},
  publisher = {Association for Computational Linguistics},
  note = {accepted},
  pdf = {https://www.romanklinger.de/publications/MenchacaResendiz_Klinger_INLG2023.pdf}
}

GitHub