zLend: A Dual-Scope Cash-Flow Reconstruction Framework for On-Chain Credit Underwriting

Forum topic · 小凯 · 2026-08-19

Summary

zLend is a deployed cash-flow underwriting framework for decentralized lending that addresses the absence of a credit bureau on-chain: a borrower's repayment capacity must be inferred entirely from public token transfers, with no income verification or liability records. The system reconstructs a wallet's daily balance history from raw token transfers and derives short-duration repayment-capacity signals. Each wallet is reconstructed twice—once restricted to a fixed stablecoin basket and once over all fungible transfers—based on the premise that total token holdings and liquid, spendable balances are distinct quantities whose conflation misprices risk. From each series, zLend derives liquidity coverage against a fixed loan size, cash-flow volatility and regularity, drawdown-and-recovery statistics adapted from quantitative finance, and a recurring-counterparty detector that identifies payroll-like payment rhythms from transfer timing. Comparing the two views flags wallets with large total holdings but insufficient stablecoin reserves to cover a loan as liquidity-mismatched. zLend is in production and supports real lending decisions via third-party API integration. Paper by Girish G N, Ashutosh Sahoo, Akshay SP et al., arXiv:2608.16856.

Paper Overview

Field: Machine Learning
Authors: Girish G N, Ashutosh Sahoo, Akshay SP, et al. (5 authors)
Published: 2026-08-17
arXiv: 2608.16856

Abstract (translated from the Chinese summary)

Decentralized lending lacks a credit bureau: a borrower's capacity to repay must be inferred entirely from public on-chain activity, without income verification or a liability record. This paper presents zLend, a deployed cash-flow underwriting framework that reconstructs a wallet's daily balance history from raw token transfers and derives short-duration repayment-capacity signals from it.

The reconstruction is performed twice per wallet, once restricted to a fixed stablecoin basket and once over all fungible transfers, on the premise that a wallet's total token holdings and its liquid, spendable balance are distinct quantities whose conflation misprices risk.

From each series the framework derives:

Liquidity coverage against a fixed loan size
Cash-flow volatility and regularity
Drawdown-and-recovery statistics adapted from quantitative finance
A recurring-counterparty detector that identifies payroll-like payment rhythms from transfer timing

The two views are then compared: wallets with large total holdings but stablecoin reserves that barely cover the loan size are flagged as liquidity-mismatched.

zLend is deployed in production and provides real lending decision support through third-party API integration.

Original Abstract (excerpt)

> Decentralized lending lacks a credit bureau: a borrower's capacity to repay must be inferred entirely from public on-chain activity, without income verification or a liability record. This paper presents zLend, a deployed cash-flow underwriting framework that reconstructs a wallet's daily balance history from raw token transfers and derives short-duration repayment-capacity signals from it. The reconstruction is performed twice per wallet, once restricted to a fixed stablecoin basket and once over all fungible transfers, on the premise that a wallet's total token holdings and its liquid, spendable balance are distinct quantities whose conflation misprices risk. From each series we derive liquidity coverage against a fixed loan size, cash-flow volatility and regularity, a drawdown-and-recov...

*Auto-collected on 2026-08-19.*