CASPER: Contrastive Approach for Smart Ponzi Scheme Detecter with More Negative Samples

2025-07-18

AI Summary

To address the performance bottleneck of intelligent Ponzi scheme detection on blockchain caused by scarce labeled data, this paper proposes a contrastive learning-based self-supervised detection framework. The method integrates graph neural networks with a multi-source negative sample augmentation strategy to extract robust deep semantic features from smart contract code, thereby significantly improving the utilization efficiency of unlabeled data. Its key innovation lies in a fine-grained negative sample construction mechanism that enhances semantic discriminability and reduces reliance on labeled data. Evaluated on the XBlock dataset, the framework achieves an F1-score improvement of nearly 20% over baseline methods when trained with only 25% of the labeled data, and yields a 2.3% F1 gain using the full labeled set—outperforming existing approaches. These results demonstrate its strong generalizability and practical effectiveness for real-world deployment.

Technology Category

• Machine Learning: Unsupervised & Self-Supervised Learning
• Search and Optimization: Learning to Search
• Natural Language Processing: Fact-Checking / Misinformation Detection (NLP Focus)

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Security and Privacy: Cryptocurrency and smart contracts
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

The rapid evolution of digital currency trading, fueled by the integration of blockchain technology, has led to both innovation and the emergence of smart Ponzi schemes. A smart Ponzi scheme is a fraudulent investment operation in smart contract that uses funds from new investors to pay returns to earlier investors. Traditional Ponzi scheme detection methods based on deep learning typically rely on fully supervised models, which require large amounts of labeled data. However, such data is often scarce, hindering effective model training. To address this challenge, we propose a novel contrastive learning framework, CASPER (Contrastive Approach for Smart Ponzi detectER with more negative samples), designed to enhance smart Ponzi scheme detection in blockchain transactions. By leveraging contrastive learning techniques, CASPER can learn more effective representations of smart contract source code using unlabeled datasets, significantly reducing both operational costs and system complexity. We evaluate CASPER on the XBlock dataset, where it outperforms the baseline by 2.3% in F1 score when trained with 100% labeled data. More impressively, with only 25% labeled data, CASPER achieves an F1 score nearly 20% higher than the baseline under identical experimental conditions. These results highlight CASPER's potential for effective and cost-efficient detection of smart Ponzi schemes, paving the way for scalable fraud detection solutions in the future.

Problem

• Detect smart Ponzi schemes in blockchain transactions
• Reduce reliance on labeled data for model training
• Improve detection accuracy with limited labeled data

Innovation

• Contrastive learning for smart Ponzi detection
• Uses unlabeled data to reduce costs
• Improves F1 score with fewer labels