Abstract
Persuasion techniques detection in news in a multi-lingual setup is non-trivial and comes with challenges, including little training data. Our system successfully leverages (back-)translation as data augmentation strategies with multi-lingual transformer models for the task of detecting persuasion techniques. The automatic and human evaluation of our augmented data allows us to explore whether (back-)translation aid or hinder performance. Our in-depth analyses indicate that both data augmentation strategies boost performance; however, balancing human-produced and machine-generated data seems to be crucial.- Anthology ID:
- 2023.semeval-1.198
- Volume:
- Proceedings of the 17th International Workshop on Semantic Evaluation (SemEval-2023)
- Month:
- July
- Year:
- 2023
- Address:
- Toronto, Canada
- Editors:
- Atul Kr. Ojha, A. Seza Doğruöz, Giovanni Da San Martino, Harish Tayyar Madabushi, Ritesh Kumar, Elisa Sartori
- Venue:
- SemEval
- SIG:
- SIGLEX
- Publisher:
- Association for Computational Linguistics
- Note:
- Pages:
- 1433–1446
- Language:
- URL:
- https://aclanthology.org/2023.semeval-1.198
- DOI:
- 10.18653/v1/2023.semeval-1.198
- Cite (ACL):
- Neele Falk, Annerose Eichel, and Prisca Piccirilli. 2023. NAP at SemEval-2023 Task 3: Is Less Really More? (Back-)Translation as Data Augmentation Strategies for Detecting Persuasion Techniques. In Proceedings of the 17th International Workshop on Semantic Evaluation (SemEval-2023), pages 1433–1446, Toronto, Canada. Association for Computational Linguistics.
- Cite (Informal):
- NAP at SemEval-2023 Task 3: Is Less Really More? (Back-)Translation as Data Augmentation Strategies for Detecting Persuasion Techniques (Falk et al., SemEval 2023)
- PDF:
- https://preview.aclanthology.org/ingest-acl-2023-videos/2023.semeval-1.198.pdf