Learning to Fuse: Cost-Sensitive Credit Assessment via Hierarchical Multi-Agent Reinforcement Learning
Preprint
- 9 May 2026
- preprint
- Published by MDPI AG in Preprints.org
Abstract
Credit risk assessment requires both accurate prediction and structured decomposition of how hetero-geneous evidence contributes to each decision. Monolithic Large Language Models can incorporate unstructured evidence and natural-language reasoning into such workflows, but in high-stakes un- derwriting they may be distracted by noisy inputs, miss rare but decisive risk cues, and offer limited control over policy-dependent decision thresholds. We present CREDITAGENT, a hierarchical credit review system with three stages: evidence filtering, specialist risk analysis by agents, and decision fusion. Our central contribution is holding the adapted backbone, specialist-agent outputs, hard-stop rules, and data split fixed, we vary only the final fusion strategy to isolate the effect of hierarchical fusion on underwriting quality. On a held-out set of 6,000 personal credit cases from Chinese financial institution, CREDITAGENT achieves 83.32% accuracy and a Business Efficiency Coefficient of 0.7647 outperform flagship model. We present these findings as an institution-specific case study while identifying which components (hierarchical fusion, GRPO training recipe) are mechanism-portable versus institution-specific (hard-stop rules, cost ratios). To ensure reproducibility, we make code and dataset publicly available at https://github.com/kouzhizhuo/Credit_Agents.Keywords
All Related Versions
This publication has 40 references indexed in Scilit:
- An analytical approach to credit risk assessment using machine learning modelsDecision Analytics Journal, 2025
- Mitigating reasoning hallucination through Multi-agent Collaborative FilteringExpert Systems with Applications, 2024
- Fraud Detection Using Machine Learning and Deep LearningSN Computer Science, 2024
- Bayesian learning models to measure the relative impact of ESG factors on credit ratingsInternational Journal of Data Science and Analytics, 2023
- Loan default prediction using a credit rating-specific and multi-objective ensemble learning schemeInformation Sciences, 2023
- The Use of Responsible Artificial Intelligence Techniques in the Context of Loan Approval ProcessesInternational Journal of Human–Computer Interaction, 2022
- The determinants of banks' credit risk: Review of the literature and future research agendaInternational Journal of Finance & Economics, 2020
- Machine learning techniques for credit risk evaluation: a systematic literature reviewJournal of Banking and Financial Technology, 2020
- A Comparative Assessment of Credit Risk Model Based on Machine Learning ——a case study of bank loan dataProcedia Computer Science, 2020
- Reassessment and Monitoring of Loan Applications with Machine LearningApplied Artificial Intelligence, 2018