PIXELMOD: Improving Soft Moderation of Visual Misleading Information on Twitter
Published in 33rd Usenix Security Symposium, 2024
⏳⏳ Pre-print coming soon. Stay tuned! 👀
Images are a powerful and immediate vehicle for online information, and are increasingly used to carry misleading or outright false messages. While online platforms like Twitter started adding soft moderation labels to image content, identifying image-based misinformation at scale poses unique challenges. In this paper, we present PixelMod, a system that leverages perceptual hashes, vector databases, and optical character recognition (OCR) to efficiently identify images that are candidates for soft moderation on Twitter. We show that PixelMod outperforms existing image similarity approaches when applied to soft moderation, overcoming their fundamental limitations, with negligible performance overhead. We then test PixelMod on a dataset of tweets surrounding the 2020 US Presidential Election, and find that it is able to identify visually misleading images that are candidates for soft moderation with 0.99% false detection and 2.06% false negatives.
Recommended citation: P. Paudel, C. Ling, J.Blackburn and G. Stringhini, “PIXELMOD: Improving Soft Moderation of Visual Misleading Information on Twitter,” 33rd Usenix Security Symposium, Philadelphia PA, USA, 2024