This paper proposes a modular and explainable architecture for Internet Meme classification and understanding. We design and implement multimodal classification ...

Dec 11, 2022 · In this paper, we pursue a modular and explainable architecture for Internet meme understanding. We design and implement multimodal ...

In this paper, we pursue a modular and explainable architecture for Internet meme understanding. We design and implement multimodal classification methods that ...

We use example- based classification because it helps users to understand how the classification model represents a meme compared to the training dataset ...

A modular and explainable architecture for Internet meme understanding that performs example- and prototype-based reasoning over training cases, ...

Recent studies [4,18] have shown that people sometimes use memes to disseminate violent. [12], hateful [6], and misogynistic messages disguised as humor [6,15].

These three examples serve to show how the dual classification system functions to categorize memes and to help us explain how they make new meanings. 4.

Jan 24, 2024 · In this paper, we propose an explainable approach to detect harmful memes, achieved through reasoning over conflicting rationales from both harmless and ...

We study the relevance of our modular and explainable models in detecting harmful memes on two existing tasks: Hate Speech Detection and Misogyny Classification ...