Large Language Models for Multi-label Propaganda Detection
This work addresses the societal problem of online propaganda spread, but it is incremental as it focuses on a specific shared task with modest performance gains.
The paper tackled multi-label propaganda detection by identifying 21 propaganda techniques in text, achieving a micro-F1 score of 59.73% using an ensemble of five models.
The spread of propaganda through the internet has increased drastically over the past years. Lately, propaganda detection has started gaining importance because of the negative impact it has on society. In this work, we describe our approach for the WANLP 2022 shared task which handles the task of propaganda detection in a multi-label setting. The task demands the model to label the given text as having one or more types of propaganda techniques. There are a total of 21 propaganda techniques to be detected. We show that an ensemble of five models performs the best on the task, scoring a micro-F1 score of 59.73%. We also conduct comprehensive ablations and propose various future directions for this work.