PIXELMOD: Improving Soft Moderation of Visual Misleading Information on Twitter

TL;DR

PIXELMOD is a system that uses perceptual hashes, vector databases, and OCR to efficiently detect visually misleading images on Twitter, improving soft moderation with high accuracy and minimal performance impact.

Contribution

The paper introduces PIXELMOD, a novel system combining perceptual hashes, vector databases, and OCR for scalable image misinformation detection on social media.

Findings

Outperforms existing image similarity methods in soft moderation tasks.

Achieves 0.99% false detection rate and 2.06% false negatives on election-related tweets.

Operates with negligible performance overhead.

Abstract

Images are a powerful and immediate vehicle to carry misleading or outright false messages, yet 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 to receive soft moderation labels on Twitter. We show that PIXELMOD outperforms existing image similarity approaches when applied to soft moderation, 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.