The age of social media is rife with memes. Understanding and detecting harmful memes pose a significant challenge due to their implicit meaning that is not explicitly conveyed through the surface text and image. However, existing harmful meme detection approaches only recognize superficial harm-indicative signals in an end-to-end classification manner but ignore in-depth cognition of the meme text and image. In this paper, we attempt to detect harmful memes based on advanced reasoning over the interplay of multimodal information in memes. Inspired by the success of Large Language Models (LLMs) on complex reasoning, we first conduct abductive reasoning with LLMs. Then we propose a novel generative framework to learn reasonable thoughts from LLMs for better multimodal fusion and lightweight fine-tuning, which consists of two training stages: 1) Distill multimodal reasoning knowledge from LLMs; and 2) Fine-tune the generative framework to infer harmfulness. Extensive experiments conducted on three meme datasets demonstrate that our proposed approach achieves superior performance than state-of-the-art methods on the harmful meme detection task.
@article{arxiv.2312.05434,
title = {Beneath the Surface: Unveiling Harmful Memes with Multimodal Reasoning Distilled from Large Language Models},
author = {Hongzhan Lin and Ziyang Luo and Jing Ma and Long Chen},
journal= {arXiv preprint arXiv:2312.05434},
year = {2023}
}
Comments
The first work to alleviate the issue of superficial understanding for harmful meme detection by explicitly utilizing commonsense knowledge, from a fresh perspective on harnessing advanced Large Language Models