I know what you MEME! Understanding and Detecting Harmful Memes with Multimodal Large Language Models
Yong Zhuang
Network and Distributed System Security (NDSS) Symposium 2025 · Day 1 · LLM Security
In an era dominated by digital communication, **memes** have emerged as a pervasive and powerful form of expression, blending images and text to convey ideas, humor, and narratives across social media platforms. While often a source of lighthearted entertainment, memes possess a significant "dark side," readily exploited by malicious actors to spread harmful content. This talk, presented by Yong Zhuang from the University at Buffalo at the NDSS Symposium, delves into the intricate challenges of understanding and detecting these detrimental memes, proposing a novel approach leveraging **Multimodal Large Language Models (MLLMs)**.
AI review
Legitimate academic research on a real problem — multimodal harmful meme detection using chain-of-thought prompted MLLMs — with solid empirical results and a clear problem decomposition. Competent work, but the core contribution is prompt engineering on top of GPT-4, which puts a ceiling on how novel this actually is.