Argumentative Stance Prediction: An Exploratory Study on Multimodality and Few-Shot Learning
This work addresses stance prediction for social media analysis, but it is incremental as it primarily compares existing methods on a specific dataset.
The study tackled argumentative stance prediction in tweets by comparing multimodal and few-shot learning approaches, finding that an ensemble of fine-tuned text-based models achieved an F1-score of 0.817, outperforming multimodal models at 0.677 and few-shot LLMs at 0.550.
To advance argumentative stance prediction as a multimodal problem, the First Shared Task in Multimodal Argument Mining hosted stance prediction in crucial social topics of gun control and abortion. Our exploratory study attempts to evaluate the necessity of images for stance prediction in tweets and compare out-of-the-box text-based large-language models (LLM) in few-shot settings against fine-tuned unimodal and multimodal models. Our work suggests an ensemble of fine-tuned text-based language models (0.817 F1-score) outperforms both the multimodal (0.677 F1-score) and text-based few-shot prediction using a recent state-of-the-art LLM (0.550 F1-score). In addition to the differences in performance, our findings suggest that the multimodal models tend to perform better when image content is summarized as natural language over their native pixel structure and, using in-context examples improves few-shot performance of LLMs.