Early Rug Pull Warning for BSC Meme Tokens via Multi-Granularity Wash-Trading Pattern Profiling

2026-03-14

AI Summary

This study addresses the challenge of effectively detecting “rug pull” risks in meme tokens on the Binance Smart Chain (BSC), which are exacerbated by high-frequency token launches and short-term speculation. Existing approaches struggle due to label sparsity, anomaly rarity, and insufficient interpretability. To overcome these limitations, this work proposes an end-to-end early-warning framework that integrates multi-granular signals—including transaction, address, and fund flow data—and constructs a 12-dimensional token-level behavioral feature set based on three wash-trading patterns: Self, Matched, and Circular. Using supervised models such as Random Forest, the framework achieves strong predictive performance on a dataset of seven tokens and 33,242 records, yielding an AUC of 0.9098, PR-AUC of 0.9185, and F1-score of 0.7429, with an average early warning lead time of 3.81 hours, thereby demonstrating the efficacy and deployability of weakly supervised, multi-granular wash-trading features.

Technology Category

• Data Mining & Knowledge Management: Anomaly/Outlier Detection
• Machine Learning: Online Learning & Bandits
• Natural Language Processing: Fact-Checking / Misinformation Detection (NLP Focus)

Application Category

• Web Mining and Content Analysis: Bridging structured and unstructured data
• Security and Privacy: Cryptocurrency and smart contracts
• Economics, Online Markets and Human Computation: Economic aspects of blockchain and cryptocurrencies

Abstract

The high-frequency issuance and short-cycle speculation of meme tokens in decentralized finance (DeFi) have significantly amplified rug-pull risk. Existing approaches still struggle to provide stable early warning under scarce anomalies, incomplete labels, and limited interpretability. To address this issue, an end-to-end warning framework is proposed for BSC meme tokens, consisting of four stages: dataset construction and labeling, wash-trading pattern feature modeling, risk prediction, and error analysis. Methodologically, 12 token-level behavioral features are constructed based on three wash-trading patterns (Self, Matched, and Circular), unifying transaction-, address-, and flow-level signals into risk vectors. Supervised models are then employed to output warning scores and alert decisions. Under the current setting (7 tokens, 33,242 records), Random Forest outperforms Logistic Regression on core metrics, achieving AUC=0.9098, PR-AUC=0.9185, and F1=0.7429. Ablation results show that trade-level features are the primary performance driver (Delta PR-AUC=-0.1843 when removed), while address-level features provide stable complementary gain (Delta PR-AUC=-0.0573). The model also demonstrates actionable early-warning potential for a subset of samples, with a mean Lead Time (v1) of 3.8133 hours. The error profile (FP=1, FN=8) indicates that the current system is better positioned as a high-precision screener rather than a high-recall automatic alarm engine. The main contributions are threefold: an executable and reproducible rug-pull warning pipeline, empirical validation of multi-granularity wash-trading features under weak supervision, and deployment-oriented evidence through lead-time and error-bound analysis.

Problem

• rug pull
• meme tokens
• wash trading
• early warning
• DeFi

Innovation

• wash-trading pattern
• multi-granularity feature modeling
• rug-pull early warning
• weak supervision
• DeFi security