MemeLens: Multilingual Multitask VLMs for Memes

TL;DR

MemeLens is a multilingual, multitask vision-language model that unifies 38 meme datasets to improve understanding of memes across various tasks and languages.

Contribution

It consolidates diverse meme datasets into a shared taxonomy and provides a comprehensive analysis of modeling paradigms and dataset interactions.

Findings

Multimodal training is essential for robust meme understanding.

Models vary significantly across semantic categories.

Fine-tuning on individual datasets can lead to over-specialization.

Abstract

Memes are a dominant medium for online communication and manipulation because meaning emerges from interactions between embedded text, imagery, and cultural context. Existing meme research is distributed across tasks (hate, misogyny, propaganda, sentiment, humour) and languages, which limits cross-domain generalization. To address this gap we propose MemeLens, a unified multilingual and multitask explanation-enhanced Vision Language Model (VLM) for meme understanding. We consolidate $38$ public meme datasets, filter and map dataset-specific labels into a shared taxonomy of $20$ tasks spanning harm, targets, figurative/pragmatic intent, and affect. We present a comprehensive empirical analysis across modeling paradigms, task categories, and datasets. Our findings suggest that robust meme understanding requires multimodal training, exhibits substantial variation across semantic categories, and remains sensitive to over-specialization when models are fine-tuned on individual datasets rather than trained in a unified setting. We make the experimental resources (https://github.com/MohamedBayan/MemeLens), model (https://huggingface.co/QCRI/MemeLens-VLM) and datasets (https://huggingface.co/datasets/QCRI/MemeLens) publicly available to the community.