Nosy Layers, Noisy Fixes: Tackling DRAs in Federated Learning Systems Using Explainable AI

Federated Learning FL has emerged as a powerful paradigm for collaborative model training while keeping client data decentralized and private. However, it is vulnerable to Data Reconstruction Attacks DRA such as "LoKI" and "Robbing the Fed", where malicious models sent from the server to the clie...

Unredacting Pixelated Text

Experiments in unredacting text that has been pixelated...

This New Tool Can Retrieve Pixelated Text from Redacted Documents

The practice of blurring out text using a method called pixelation may not be as secure as previously thought. While the most foolproof way of concealing sensitive textual information is to use opaque black bars, other redaction methods like pixelation can achieve the opposite effect, enabling th...