Beneath the Surface: Unveiling Harmful Memes with Multimodal Reasoning Distilled from Large Language Models
Citations
0
Open access
No
Source
arxiv
OpenAlex
Not enriched
arXiv
2312.05434
Abstract
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.
Collections
Add to collection
Paper intelligence
Analyse this paper
Extract the research problem, methodology, findings, contributions, limitations, datasets, equations and future research directions.
No graph connections yet.
Sync citations or add papers to shared collections to build this network.
Knowledge graph
Citation network
Explore references, papers that cite this work and related papers in your Codex library.
References
0No references have been linked yet.
Cited by
0No saved paper is currently linked as citing this work.
Related papers
0Add papers to shared collections or enrich their topics to find related work.
Research workspace
Attach the paper PDF, extract its text, classify its contents and create semantic embeddings.
Paper resources
No PDF assets have been attached to this paper yet.