MemeCLIP: Leveraging CLIP Representations for Multimodal Meme Classification

TL;DR

MemeCLIP leverages CLIP representations to improve multimodal meme classification across multiple linguistic aspects, introducing a new dataset and achieving superior results over existing methods.

Contribution

This paper presents MemeCLIP, a novel framework that enhances multimodal meme classification by utilizing pre-trained CLIP, along with a new dataset for LGBTQ+ Pride memes.

Findings

MemeCLIP outperforms previous models on benchmark datasets.

The new PrideMM dataset fills a gap in multimodal meme analysis.

MemeCLIP shows competitive zero-shot performance compared to GPT-4.

Abstract

The complexity of text-embedded images presents a formidable challenge in machine learning given the need for multimodal understanding of multiple aspects of expression conveyed by them. While previous research in multimodal analysis has primarily focused on singular aspects such as hate speech and its subclasses, this study expands this focus to encompass multiple aspects of linguistics: hate, targets of hate, stance, and humor. We introduce a novel dataset PrideMM comprising 5,063 text-embedded images associated with the LGBTQ+ Pride movement, thereby addressing a serious gap in existing resources. We conduct extensive experimentation on PrideMM by using unimodal and multimodal baseline methods to establish benchmarks for each task. Additionally, we propose a novel framework MemeCLIP for efficient downstream learning while preserving the knowledge of the pre-trained CLIP model. The results of our experiments show that MemeCLIP achieves superior performance compared to previously proposed frameworks on two real-world datasets. We further compare the performance of MemeCLIP and zero-shot GPT-4 on the hate classification task. Finally, we discuss the shortcomings of our model by qualitatively analyzing misclassified samples. Our code and dataset are publicly available at: https://github.com/SiddhantBikram/MemeCLIP.