Memes in the Wild: Assessing the Generalizability of the Hateful Memes Challenge Dataset
This provides a reality check for hateful meme detection benchmarks, highlighting limitations for real-world applications, though it is incremental as it focuses on dataset evaluation rather than new methods.
The study evaluated whether models trained on the Facebook Hateful Memes Challenge dataset generalize to real-world memes from Pinterest, finding that out-of-sample performance drops due to OCR noise and greater diversity in meme formats.
Hateful memes pose a unique challenge for current machine learning systems because their message is derived from both text- and visual-modalities. To this effect, Facebook released the Hateful Memes Challenge, a dataset of memes with pre-extracted text captions, but it is unclear whether these synthetic examples generalize to `memes in the wild'. In this paper, we collect hateful and non-hateful memes from Pinterest to evaluate out-of-sample performance on models pre-trained on the Facebook dataset. We find that memes in the wild differ in two key aspects: 1) Captions must be extracted via OCR, injecting noise and diminishing performance of multimodal models, and 2) Memes are more diverse than `traditional memes', including screenshots of conversations or text on a plain background. This paper thus serves as a reality check for the current benchmark of hateful meme detection and its applicability for detecting real world hate.