Detecting rug pulls in decentralized exchanges: The rise of meme coins

Выявление схем rug pull на децентрализованных биржах: рост мем-коинов
Yury Yanovich, Igor Vodolazov, Pavel Kuznetsov, Alisa Kalacheva
2025-07-15

Ethereum blockchaindecentralized exchangesmachine learningmeme coinsrug pull detection
The surge in cryptoasset valuations and the expansive landscape for their inception have spurred a surge in illicit activities within the market. Decentralized exchanges (DEX) have facilitated trading with a wide array of tokens, even in scenarios of limited liquidity. Fraudulent practices manifest diversely, encompassing counterfeit tokens, rug pulls, and pump-and-dump schemes, all sharing a common thread of lacking functional innovation and relying heavily on aggressive social media and messenger marketing tactics. This study represents a critical stride towards the identification and profiling of deceitful tokens on DEX platforms. Our approach involved compiling data on new tokens spanning multiple years from the Ethereum blockchain, alongside tracking all associated purchase and sale transactions. Our analysis revealed that Uniswap V2 predominantly hosts the trading of new tokens, with an alarming discovery that over 98% of tokens minted daily exhibit fraudulent characteristics. Subsequently, a machine learning model was constructed to forecast the likelihood of a rug pull occurring shortly after trading commencement. The findings underscore the imperative nature of identifying fraudulent activities and stress the necessity for collaborative efforts between decentralized exchanges and regulatory bodies to mitigate financial losses experienced by investors.
1
A machine-learning model was developed to predict whether a rug pull will occur shortly after trading begins.
2
More than 98% of tokens minted daily were identified as exhibiting fraudulent characteristics.
3
The findings emphasize collaboration between decentralized exchanges and regulators to reduce investor financial losses.
4
The study compiles multi-year data on newly created Ethereum tokens and their associated purchase and sale transactions.
5
Uniswap V2 predominantly hosts trading of newly created tokens across decentralized exchanges.

Newly minted tokens traded on Ethereum-based decentralized exchanges, particularly Uniswap V2

Fraudulent-token characteristics and the prediction of rug pulls shortly after trading begins

Publication Details
Publication Date
2025-07-15
Journal
Publisher
ISSN
Cited by
4
Access Type
Author Information
Authors
Yury Yanovich
Igor Vodolazov
Pavel Kuznetsov
Alisa Kalacheva
Explore further
Open the scid.ai AI chat with a ready-made request: it will find papers on a similar topic and help build a literature review.
Find similar papers in the chat →
Make a presentation
100%