Correlating Account on Ethereum Mixing Service via Domain-Invariant feature learning

2025-05-15

AI Summary

To address the transaction traceability challenge posed by Ethereum mixers (e.g., Tornado Cash), existing account linkage methods suffer from scarce labeled data and sensitivity to label noise. This paper proposes StealthLink, the first framework to transfer knowledge from blockchain anomaly detection to mixer tracing via cross-task knowledge transfer, enabling domain-invariant feature learning. Its core contributions are: (1) MixFusion, a subgraph encoding mechanism that jointly models multi-hop transaction topology and mixer semantics; and (2) an adversarial discrepancy minimization strategy that aligns feature distributions between the source domain (anomaly detection) and target domain (mixer linkage). Evaluated on real-world mixer data, StealthLink achieves 96.98% F1-score with only 10 labeled samples—substantially outperforming supervised baselines—while demonstrating strong robustness to class imbalance and label noise.

Technology Category

• Machine Learning: Mixture of Experts (MoE)
• Data Mining & Knowledge Management: Anomaly/Outlier Detection
• Search and Optimization: Mixed Discrete/Continuous Search

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Security and Privacy: Data transparency and provenance

Abstract

The untraceability of transactions facilitated by Ethereum mixing services like Tornado Cash poses significant challenges to blockchain security and financial regulation. Existing methods for correlating mixing accounts suffer from limited labeled data and vulnerability to noisy annotations, which restrict their practical applicability. In this paper, we propose StealthLink, a novel framework that addresses these limitations through cross-task domain-invariant feature learning. Our key innovation lies in transferring knowledge from the well-studied domain of blockchain anomaly detection to the data-scarce task of mixing transaction tracing. Specifically, we design a MixFusion module that constructs and encodes mixing subgraphs to capture local transactional patterns, while introducing a knowledge transfer mechanism that aligns discriminative features across domains through adversarial discrepancy minimization. This dual approach enables robust feature learning under label scarcity and distribution shifts. Extensive experiments on real-world mixing transaction datasets demonstrate that StealthLink achieves state-of-the-art performance, with 96.98% F1-score in 10-shot learning scenarios. Notably, our framework shows superior generalization capability in imbalanced data conditions than conventional supervised methods. This work establishes the first systematic approach for cross-domain knowledge transfer in blockchain forensics, providing a practical solution for combating privacy-enhanced financial crimes in decentralized ecosystems.

Problem

• Enhancing traceability of Ethereum mixing transactions
• Overcoming data scarcity in mixing account correlation
• Transferring anomaly detection knowledge to transaction tracing

Innovation

• Cross-task domain-invariant feature learning
• MixFusion module encodes mixing subgraphs
• Adversarial discrepancy minimization aligns features