A prompt-based pipeline with GPT-4o detects hateful memes at state-of-the-art zero-shot accuracy and mitigates them by replacing hateful text or images, with 88% of 631 human-rated mitigated memes judged non-hateful.
Recent advances in online hate speech moderation: Multimodality and the role of large models
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Detecting and Mitigating Hateful Content in Multimodal Memes with Vision-Language Models
A prompt-based pipeline with GPT-4o detects hateful memes at state-of-the-art zero-shot accuracy and mitigates them by replacing hateful text or images, with 88% of 631 human-rated mitigated memes judged non-hateful.