Adversarial Training for News Stance Detection: Leveraging Signals from a Multi-Genre Corpus.
View / Open Files
Authors
Conforti, C
Berndt, J
Pilehvar, MT
Basaldella, M
Giannitsarou, C
Toxvaerd, F
Collier, N
Publication Date
2021-01-01Journal Title
EACL Hackashop on News Media Content Analysis and Automated Report Generation, Hackashop 2021 at 16th conference of the European Chapter of the Association for Computational Linguistics, EACL 2021 - Proceedings
Conference Name
Hackashop 2021 at 16th conference of the European Chapter of the Association for Computational Linguistics, EACL 2021 - Proceedings
ISBN
9781954085138
Publisher
Association for Computational Linguistics
Pages
1-7
Type
Conference Object
This Version
VoR
Metadata
Show full item recordCitation
Conforti, C., Berndt, J., Pilehvar, M., Basaldella, M., Giannitsarou, C., Toxvaerd, F., & Collier, N. (2021). Adversarial Training for News Stance Detection: Leveraging Signals from a Multi-Genre Corpus.. EACL Hackashop on News Media Content Analysis and Automated Report Generation, Hackashop 2021 at 16th conference of the European Chapter of the Association for Computational Linguistics, EACL 2021 - Proceedings, 1-7. https://doi.org/10.17863/CAM.84535
Abstract
Cross-target generalization constitutes an important issue for news Stance Detection (SD). In this short paper, we investigate adversarial cross-genre SD, where knowledge from annotated user-generated data is leveraged to improve news SD on targets unseen during training. We implement a BERT-based adversarial network and show experimental performance improvements over a set of strong baselines. Given the abundance of user-generated data, which are considerably less expensive to retrieve and annotate than news articles, this constitutes a promising research direction.
Identifiers
External DOI: https://doi.org/10.17863/CAM.84535
This record's URL: https://www.repository.cam.ac.uk/handle/1810/337116
Statistics
Total file downloads (since January 2020). For more information on metrics see the
IRUS guide.