Explain First, Trust Later: LLM-Augmented Explanations for Graph-Based Crypto Anomaly Detection

TL;DR

This paper proposes a novel approach that combines explanations and trust-building for graph-based crypto anomaly detection using large language models, aiming to improve detection accuracy and interpretability in DeFi security.

Contribution

It introduces a new LLM-augmented explanation framework that enhances trust and understanding in crypto anomaly detection methods, addressing the challenge of detecting financial crimes in DeFi.

Findings

Improved detection accuracy over baseline models

Enhanced interpretability of anomaly detection results

Increased user trust through explainability

Abstract

The decentralized finance (DeFi) community has grown rapidly in recent years, pushed forward by cryptocurrency enthusiasts interested in the vast untapped potential of new markets. The surge in popularity of cryptocurrency has ushered in a new era of financial crime. Unfortunately, the novelty of the technology makes the task of catching and prosecuting offenders particularly challenging. Thus, it is necessary to implement automated detection tools related to policies to address the growing criminality in the cryptocurrency realm.