Detecting Rug Pulls in Decentralized Exchanges: Machine Learning Evidence from the TON Blockchain

2025-09-01

AI Summary

This work addresses the early detection of rug pull scams in TON blockchain-based decentralized exchanges (DEXs). We propose the first machine learning–based early-warning framework designed for multiple platforms—Ston.Fi and DeDust. Methodologically, we introduce the first unified model comparing two distinct rug pull definitions—sudden TVL decline versus transaction stagnation—revealing cross-platform disparities in feature distributions and enabling platform-adaptive modeling. Our approach integrates asynchronous on-chain transaction data, liquidity metrics, and temporal features into a gradient-boosting model. Experiments demonstrate minute-level detection within five minutes of scam initiation: under the TVL-decline definition, the model achieves an AUC of 0.891; under the transaction-stagnation definition, it attains significantly higher recall. This work delivers both a transferable methodology and a practical early-warning tool for fraud detection in multi-chain DeFi ecosystems.

Technology Category

• Data Mining & Knowledge Management: Anomaly/Outlier Detection
• Machine Learning: Other Foundations of Machine Learning
• Reasoning under Uncertainty: Other Foundations of Reasoning under Uncertainty

Application Category

• Economics, Online Markets and Human Computation: LLM based quality controls for crowd work
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Web Mining and Content Analysis: Large pretrained models with web data

Abstract

This paper presents a machine learning framework for the early detection of rug pull scams on decentralized exchanges (DEXs) within The Open Network (TON) blockchain. TON's unique architecture, characterized by asynchronous execution and a massive web2 user base from Telegram, presents a novel and critical environment for fraud analysis. We conduct a comprehensive study on the two largest TON DEXs, Ston.Fi and DeDust, fusing data from both platforms to train our models. A key contribution is the implementation and comparative analysis of two distinct rug pull definitions--TVL-based (a catastrophic liquidity withdrawal) and idle-based (a sudden cessation of all trading activity)--within a single, unified study. We demonstrate that Gradient Boosting models can effectively identify rug pulls within the first five minutes of trading, with the TVL-based method achieving superior AUC (up to 0.891) while the idle-based method excels at recall. Our analysis reveals that while feature sets are consistent across exchanges, their underlying distributions differ significantly, challenging straightforward data fusion and highlighting the need for robust, platform-aware models. This work provides a crucial early-warning mechanism for investors and enhances the security infrastructure of the rapidly growing TON DeFi ecosystem.

Problem

• Detecting rug pull scams on TON blockchain DEXs
• Comparing TVL-based and idle-based fraud definitions
• Early identification within first five minutes of trading

Innovation

• Machine learning framework for early rug pull detection
• Comparative analysis of TVL-based and idle-based definitions
• Gradient Boosting models identify scams within five minutes