[论文解读] Early Rug Pull Warning for BSC Meme Tokens via Multi-Granularity Wash-Trading Pattern Profiling

Dingding Cao, Bianbian Jiao+3 more|arXiv (Cornell University)|Mar 14, 2026

Financial Markets and Investment Strategies被引用 0

一句话总结

本论文提出一个端到端框架，利用多粒度洗牌交易模式构建代币级风险特征，并应用监督模型为 BSC Meme 代币提供早期 rug-pull 预警，在弱监督条件下随机森林的表现优于逻辑回归。

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.

研究动机与目标

• 在弱监督条件下解决高频 BSC meme 代币 rug-pull 风险检测的挑战。
• 开发一个可复现的管线，将洗牌交易模式与可执行的警报分数联系起来。
• 识别驱动预测性能的特征组（交易级、地址级、合约级）。
• 提供面向部署的指标，如领先时间和误差边界，以帮助风险分流。

提出的方法

• 从三种洗牌交易模式（自我、匹配、循环）构建 12 个代币级行为特征。
• 将交易、地址和流信号聚合为代币级风险特征向量。
• 训练监督模型（逻辑回归和随机森林）以输出警告分数和领先时间估计值。
• 使用准确率、精确率、召回率、F1、AUC、PR-AUC 及领先时间进行评估，PR-AUC 作为核心排序指标。
• 进行消融分析以评估交易级、地址级和合约级特征的贡献。

Figure 1: BSC Meme Token Rug Pull Early-warning Framework. The overall pipeline contains four stages (E1–E4): data construction and labeling, wash-trading pattern profiling, early-warning modeling, and ablation/error analysis.

实验结果

研究问题

• RQ1 多粒度洗牌交易特征是否能在弱监督下提供稳定的早期警告信号？
• RQ2 哪些特征组对预测性能贡献最大，以及它们如何影响召回和精确度？
• RQ3 在为 BSC meme 代币部署早期警告模型时，其领先时间和误差特征是什么？
• RQ4 非线性模型与线性基线在捕捉链上风险模式方面表现有何差异？

主要发现

• 随机森林在核心指标上优于逻辑回归（AUC 0.9098，PR-AUC 0.9185，F1 0.7429）。
• 交易级特征是主要性能驱动因素（移除后 PR-AUC 降低 0.1843）。
• 地址级特征提供稳定的互补增益（移除后 PR-AUC 降低 0.0573）。
• 当前合约级特征显示有限的判别力（移除后 PR-AUC 提升 0.0077）。
• 领先时间（v1）在部分案例的均值为 3.8133 小时，中位数约为 1.0331 小时。
• 误差特征显示 FP=1、FN=8，表明更倾向于高精度筛查而非高召回自动化。

Figure 2: Data Collection and Labeling Pipeline. The workflow includes export, token-wise merging, deduplication, normalization, window capping, rule-based labeling, and quality-control checks.