GreyBox at SemEval-2024 Task 4: Progressive Fine-tuning (for Multilingual Detection of Propaganda Techniques)

Nathan Roll, Calbert Graham


Abstract
We introduce a novel fine-tuning approach that effectively primes transformer-based language models to detect rhetorical and psychological techniques within internet memes. Our end-to-end system retains multilingual and task-general capacities from pretraining stages while adapting to domain intricacies using an increasingly targeted set of examples– achieving competitive rankings across English, Bulgarian, and North Macedonian. We find that our monolingual post-training regimen is sufficient to improve task performance in 17 language varieties beyond equivalent zero-shot capabilities despite English-only data. To promote further research, we release our code publicly on GitHub.
Anthology ID:
2024.semeval-1.127
Volume:
Proceedings of the 18th International Workshop on Semantic Evaluation (SemEval-2024)
Month:
June
Year:
2024
Address:
Mexico City, Mexico
Editors:
Atul Kr. Ojha, A. Seza Doğruöz, Harish Tayyar Madabushi, Giovanni Da San Martino, Sara Rosenthal, Aiala Rosá
Venue:
SemEval
SIG:
SIGLEX
Publisher:
Association for Computational Linguistics
Note:
Pages:
888–893
Language:
URL:
https://aclanthology.org/2024.semeval-1.127
DOI:
10.18653/v1/2024.semeval-1.127
Bibkey:
Cite (ACL):
Nathan Roll and Calbert Graham. 2024. GreyBox at SemEval-2024 Task 4: Progressive Fine-tuning (for Multilingual Detection of Propaganda Techniques). In Proceedings of the 18th International Workshop on Semantic Evaluation (SemEval-2024), pages 888–893, Mexico City, Mexico. Association for Computational Linguistics.
Cite (Informal):
GreyBox at SemEval-2024 Task 4: Progressive Fine-tuning (for Multilingual Detection of Propaganda Techniques) (Roll & Graham, SemEval 2024)
Copy Citation:
PDF:
https://aclanthology.org/2024.semeval-1.127.pdf
Supplementary material:
 2024.semeval-1.127.SupplementaryMaterial.txt
Supplementary material:
 2024.semeval-1.127.SupplementaryMaterial.zip