Study tests minute-scale rug-pull detection for new Solana tokens
Original title: Early Rug Pull Detection on Solana
Transfer graphs for 2,700 SolRPDS tokens already differ in size, density, and centralization within 12.5 minutes.
Researchers study whether transfer-graph structure improves early rug-pull warnings for newly launched Solana tokens, where creation is cheap and drains can finish within minutes. They sample 2,700 tokens from SolRPDS, using one-minute Birdeye histories and Solana RPC transfers, and define a rug pull by liquidity drain, price collapse, and persistent depletion over 24 hours. Fixed windows of 12.5, 35.5, and 93 minutes compare graph neural networks with matched non-graph models on the same features.
Unconfirmed: The abstract does not state which model class outperformed the other.
- All features use only information available at or before each prediction cutoff.
- Windows were chosen in advance from the published Solana rug-pull lifespan distribution.
- Early graphs differ in size, density, reciprocity, and receiver-side centralization.
- A fixed token panel separates longer observation from changes in the token population.
- Solana tokens use standard programs rather than custom per-token contracts.
From the original article
1 min read at arxiv.orgWeb3 platforms that help users discover and trade newly launched crypto tokens, including launchpads, decentralized-exchange (DEX) interfaces, wallets, and token-analytics sites, must assess tokens before their legitimacy is clear. On Solana, where token creation is inexpensive and tokens use standard programs rather than custom per-token smart contracts, rug pulls can unfold within minutes.
By Saba Heydaridoost · arxiv.org
A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.