DeXposure-FM: A Time-series, Graph Foundation Model for Credit Exposures and Stability on Decentralised Financial Networks

Credit exposure in Decentralized Finance (DeFi) is implicit and token-mediated, so shocks to widely held tokens can cascade across protocols and chains. We introduce DeXposure-FM, a time-series graph foundation model for forecasting inter-protocol credit exposure. Built on a GraphPFN-based graph-tabular encoder, DeXposure-FM is trained on 43.7M DeXposure observations covering 4,300+ protocols, 602 chains, and 24,300+ tokens. On multi-step forecasting and predictive contagion stress testing, it improves edge-existence prediction and reduces magnitude RMSE over neural baselines, while persistence remains competitive on stable magnitudes. Forecasted graphs further support macroprudential indicators, including systemic-importance scores, sector spillovers, concentration, and scenario stress losses. Gains concentrate in structural-change and tail-error regimes where carry-forward assumptions are least reliable.