arXiv cs.AI
· Papers
Beyond a Joke: Multi-Angle Reasoning for Detecting and Explaining Harmful Humor in Memes
arXiv:2607.15442v1 Announce Type: new Abstract: Internet memes intertwine visual cues, textual content, and cultural context, making them particularly challenging to interpret in scenarios where humor, sarcasm, and harmful intent coexist. These complexities highlight the need for explainable meme understanding systems