Enhancing Website Fingerprinting Attacks against Traffic Drift
Xinhao Deng
Network and Distributed System Security (NDSS) Symposium 2026 · Day 1 · Network Security
This talk introduces **Proteus**, the first adaptive website fingerprinting (WF) attack framework that continuously adapts to real-world **traffic drift** -- the systematic changes in traffic characteristics over time that cause even state-of-the-art WF attacks to fail in deployment. Unlike prior approaches that require labeled data or reference models, Proteus fine-tunes attack models using only **unlabeled traffic** observed during deployment, achieving dramatic performance recovery across six drift scenarios including temporal drift spanning nine months, Tor version changes, network condition variations across five countries, and browsing behavior drift across 17,000+ subpages.
AI review
A well-executed academic study on adapting website fingerprinting attacks to real-world traffic drift using unsupervised techniques. The three-stage pipeline is technically sound and the evaluation on 350,000+ Tor traces is impressively thorough, but this is fundamentally an ML engineering contribution rather than a novel attack technique. The adversary model remains the same -- the innovation is in maintaining attack performance over time without labeled data.