TRUST-SCF: Transformer-based Risk Understanding and Scoring for Transactional Supply Chain Finance
Researchers have developed TRUST-SCF, a novel transformer-based framework designed to enhance risk assessment in transactional supply chain finance. This system analyzes sequences of transaction data, including utilization and repayment behavior, to predict credit risk dynamically. A key innovation is its financially aligned attention bias, which allows for more nuanced comparisons of repayment patterns under similar exposure levels. Experiments on a large dataset demonstrate TRUST-SCF's effectiveness in improving delay prediction and generating credit scores that accurately reflect future repayment performance. AI
IMPACT Introduces a novel transformer-based approach for dynamic credit scoring in financial transactions.