RECTor: Robust and Efficient Correlation Attack on Tor

TL;DR

RECTor introduces a machine learning framework that significantly improves traffic correlation accuracy and efficiency in Tor, exposing vulnerabilities and emphasizing the need for stronger defenses.

Contribution

It presents a novel attention-based MIL and GRU approach for robust, scalable traffic correlation under realistic, noisy conditions.

Findings

Outperforms state-of-the-art methods by up to 60% in true positive rate.

Reduces training and inference time by over 50%.

Demonstrates near-linear scalability with increasing flow data.

Abstract

Tor is a widely used anonymity network that conceals user identities by routing traffic through encrypted relays, yet it remains vulnerable to traffic correlation attacks that deanonymize users by matching patterns in ingress and egress traffic. However, existing correlation methods suffer from two major limitations: limited robustness to noise and partial observations, and poor scalability due to computationally expensive pairwise matching. To address these challenges, we propose RECTor, a machine learning-based framework for traffic correlation under realistic conditions. RECTor employs attention-based Multiple Instance Learning (MIL) and GRU-based temporal encoding to extract robust flow representations, even when traffic data is incomplete or obfuscated. These embeddings are mapped into a shared space via a Siamese network and efficiently matched using approximate nearest neighbor (aNN) search. Empirical evaluations show that RECTor outperforms state-of-the-art baselines such as DeepCorr, DeepCOFFEA, and FlowTracker, achieving up to 60% higher true positive rates under high-noise conditions and reducing training and inference time by over 50%. Moreover, RECTor demonstrates strong scalability: inference cost grows near-linearly as the number of flows increases. These findings reveal critical vulnerabilities in Tor's anonymity model and highlight the need for advanced model-aware defenses.