Linking Cryptoasset Attribution Tags to Knowledge Graph Entities: An LLM-based Approach
This work addresses the need for more reliable forensic evidence in cryptoasset investigations, offering a domain-specific solution that is incremental in applying LLMs to a known bottleneck.
The paper tackled the problem of inconsistent or incorrect attribution tags in cryptoasset forensics by proposing a novel LLM-based method to link tags to knowledge graph entities, achieving up to 37.4% higher F1-score than baselines and 93% recall with candidate sets of five entities.
Attribution tags form the foundation of modern cryptoasset forensics. However, inconsistent or incorrect tags can mislead investigations and even result in false accusations. To address this issue, we propose a novel computational method based on Large Language Models (LLMs) to link attribution tags with well-defined knowledge graph concepts. We implemented this method in an end-to-end pipeline and conducted experiments showing that our approach outperforms baseline methods by up to 37.4% in F1-score across three publicly available attribution tag datasets. By integrating concept filtering and blocking procedures, we generate candidate sets containing five knowledge graph entities, achieving a recall of 93% without the need for labeled data. Additionally, we demonstrate that local LLM models can achieve F1-scores of 90%, comparable to remote models which achieve 94%. We also analyze the cost-performance trade-offs of various LLMs and prompt templates, showing that selecting the most cost-effective configuration can reduce costs by 90%, with only a 1% decrease in performance. Our method not only enhances attribution tag quality but also serves as a blueprint for fostering more reliable forensic evidence.