Kb at SemEval-2023 Task 3: On Multitask Hierarchical BERT Base Neural Network for Multi-label Persuasion Techniques Detection

Katarzyna Baraniak, M Sydow


Abstract
This paper presents a solution for Semeval 2023 subtask3 of task 3: persuasion techniques in paragraphs detection. The aim of this task is to identify all persuasion techniques in each paragraph of a given news article. We use hierarchical multitask neural networks combined with transformers. Span detection is an auxiliary task that helps in the main task: identifying propaganda techniques. Our experiments show that if we change the index of BERT embedding from the first token of the whole input to the first token of the identified span, it can improve performance. Span and label detection can be performed using one network, so we save data and, when data is limited, we can use more of it for training.
Anthology ID:
2023.semeval-1.193
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:
1395–1400
Language:
URL:
https://aclanthology.org/2023.semeval-1.193
DOI:
10.18653/v1/2023.semeval-1.193
Bibkey:
Cite (ACL):
Katarzyna Baraniak and M Sydow. 2023. Kb at SemEval-2023 Task 3: On Multitask Hierarchical BERT Base Neural Network for Multi-label Persuasion Techniques Detection. In Proceedings of the 17th International Workshop on Semantic Evaluation (SemEval-2023), pages 1395–1400, Toronto, Canada. Association for Computational Linguistics.
Cite (Informal):
Kb at SemEval-2023 Task 3: On Multitask Hierarchical BERT Base Neural Network for Multi-label Persuasion Techniques Detection (Baraniak & Sydow, SemEval 2023)
Copy Citation:
PDF:
https://preview.aclanthology.org/nschneid-patch-3/2023.semeval-1.193.pdf