Exploring Vision Language Models for Multimodal and Multilingual Stance Detection

TL;DR

This paper evaluates state-of-the-art Vision-Language Models on a new multilingual, multimodal stance detection dataset, revealing their reliance on text over images and their generally consistent cross-lingual predictions.

Contribution

It introduces a new dataset and provides an extensive evaluation of VLMs on multimodal and multilingual stance detection, highlighting their strengths and limitations.

Findings

VLMs rely more on text than images for stance detection.

Models tend to use text within images more than other visual cues.

Predictions are generally consistent across languages, with some outliers.

Abstract

Social media's global reach amplifies the spread of information, highlighting the need for robust Natural Language Processing tasks like stance detection across languages and modalities. Prior research predominantly focuses on text-only inputs, leaving multimodal scenarios, such as those involving both images and text, relatively underexplored. Meanwhile, the prevalence of multimodal posts has increased significantly in recent years. Although state-of-the-art Vision-Language Models (VLMs) show promise, their performance on multimodal and multilingual stance detection tasks remains largely unexamined. This paper evaluates state-of-the-art VLMs on a newly extended dataset covering seven languages and multimodal inputs, investigating their use of visual cues, language-specific performance, and cross-modality interactions. Our results show that VLMs generally rely more on text than images for stance detection and this trend persists across languages. Additionally, VLMs rely significantly more on text contained within the images than other visual content. Regarding multilinguality, the models studied tend to generate consistent predictions across languages whether they are explicitly multilingual or not, although there are outliers that are incongruous with macro F1, language support, and model size.