一句话：提出首个基于 LLM 的 DeFi 价格操纵检测方法 DeFiScope，支持标准与自定义价格模型。
DeFi 管理巨额链上资产，却易受价格操纵攻击。现有系统对自定义协议中的非标准价格模型效果有限，而这类协议占近三年 95 起相关攻击的 44.2%。本文提出首个基于 LLM 的检测方法 DeFiScope：利用 LLM 从合约源码抽象价格计算并推断代币价格趋势，借助 Foundry 合成链上数据微调价格专用 LLM，再结合从低级交易恢复的高层操作与挖掘模式检测操纵。实验显示，其对真实攻击召回率 80%，可疑交易精确率 96%，良性交易零误报，显著优于 SOTA；产业合作确认 147 起真实攻击，含 81 起新发现。
DeFi (Decentralized Finance) is one of the most important applications of today’s cryptocurrencies and smart contracts. It manages hundreds of billions in Total Value Locked (TVL) on-chain, yet it remains susceptible to common DeFi price manipulation attacks. Despite state-of-the-art (SOTA) systems like DeFiRanger and DeFort, we found that they are less effective to non-standard price models in custom DeFi protocols, which account for 44.2% of the 95 DeFi price manipulation attacks reported over the past three years.In this paper, we introduce the first LLM-based approach, DeFiScope, for detecting DeFi price manipulation attacks in both standard and custom price models. Our insight is that large language models (LLMs) have certain intelligence to abstract price calculation from smart contract source code and infer the trend of token price changes based on the extracted price models. To further strengthen LLMs in this aspect, we leverage Foundry to synthesize on-chain data and use it to fine-tune a DeFi price-specific LLM. Together with the high-level DeFi operations recovered from low-level transaction data, DeFiScope detects various DeFi price manipulations according to systematically mined patterns. Experimental results show that DeFiScope achieves a high recall of 80% on real-world attacks, a precision of 96% on suspicious transactions, and zero false alarms on benign transactions, significantly outperforming SOTA approaches. Moreover, we evaluate DeFiScope’s cost-effectiveness and demonstrate its practicality by helping our industry partner confirm 147 real-world price manipulation attacks, including discovering 81 previously unknown historical incidents.
DOI 原文 ·
@inproceedings{paperbot9404,
title = {Detecting Various DeFi Price Manipulations with LLM Reasoning},
author = {Juantao Zhong and Daoyuan Wu and Ye Liu and Maoyi Xie and Yang Liu and Yi Li and Ning Liu},
booktitle = {ASE},
year = {2025},
doi = {10.1109/ase63991.2025.00149}
}