Douwe Kiela, Hamed Firooz, Aravind Mohan, Vedanuj Goswami, Amanpreet Singh, Pratik Ringshia, Davide Testuggine
This work proposes a new challenge set for multimodal classification, focusing on detecting hate speech in multimodal memes. It is constructed such that unimodal models struggle and only multimodal models can succeed: difficult examples ("benign confounders") are added to the dataset to make it hard to rely on unimodal signals. The task requires subtle reasoning, yet is straightforward to evaluate as a binary classification problem. We provide baseline performance numbers for unimodal models, as well as for multimodal models with various degrees of sophistication. We find that state-of-the-art methods perform poorly compared to humans (64.73% vs. 84.7% accuracy), illustrating the difficulty of the task and highlighting the challenge that this important problem poses to the community.
| Task | Dataset | Metric | Value | Model |
|---|---|---|---|---|
| Meme Classification | Hateful Memes | Accuracy | 0.847 | Human |
| Meme Classification | Hateful Memes | ROC-AUC | 0.8265 | Human |
| Meme Classification | Hateful Memes | Accuracy | 0.695 | Visual BERT COCO |
| Meme Classification | Hateful Memes | ROC-AUC | 0.754 | Visual BERT COCO |