TeamAmpa at SemEval-2023 Task 3: Exploring Multilabel and Multilingual RoBERTa Models for Persuasion and Framing Detection
Amalie Brogaard Pauli, Rafael Pablos Sarabia, Leon Derczynski, Ira Assent · 2023
This paper describes our submission to the SemEval 2023 Task 3 on two subtasks: detecting persuasion techniques and framing.Both subtasks are multi-label classification problems.We present a set of experiments, exploring how to get robust performance across languages using pre-trained RoBERTa models.We test different oversampling strategies, a strategy of adding textual features from predictions obtained with related models, and present both inconclusive and negative results.We achieve a robust ranking across languages and subtasks with our best ranking being nr. 1 for Subtask 3 on Spanish.