MASCA: LLM based-Multi Agents System for Credit Assessment

2025-07-30

AI Summary

Credit assessment has long relied on rule-based and statistical models, which struggle to simultaneously ensure risk identification, profit–risk trade-off optimization, and decision fairness. To address these limitations, this paper proposes a hierarchical LLM-based multi-agent system that emulates real-world credit decision-making: a bottom-layer agent extracts heterogeneous features from multi-source data; a middle-layer agent models strategic interactions among stakeholders via signaling games; and a top-layer agent jointly optimizes the risk–return trade-off. The system further incorporates contrastive learning to enhance discriminative capability and integrates bias detection and mitigation mechanisms to improve fairness. Extensive experiments on multiple real-world credit datasets demonstrate that our approach significantly outperforms conventional models and monolithic LLM baselines—achieving AUC improvements of 3.2–5.8 percentage points. This work represents the first integration of signaling game theory with a hierarchical multi-agent architecture in financial credit scoring, establishing a novel paradigm for interpretable, robust, and fair AI-driven credit decisions.

Technology Category

• Multiagent Systems: Adversarial Agents
• Cognitive Modeling & Cognitive Systems: Agent Architectures
• Machine Learning: Learning Preferences or Rankings

Application Category

• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems

Abstract

Recent advancements in financial problem-solving have leveraged LLMs and agent-based systems, with a primary focus on trading and financial modeling. However, credit assessment remains an underexplored challenge, traditionally dependent on rule-based methods and statistical models. In this paper, we introduce MASCA, an LLM-driven multi-agent system designed to enhance credit evaluation by mirroring real-world decision-making processes. The framework employs a layered architecture where specialized LLM-based agents collaboratively tackle sub-tasks. Additionally, we integrate contrastive learning for risk and reward assessment to optimize decision-making. We further present a signaling game theory perspective on hierarchical multi-agent systems, offering theoretical insights into their structure and interactions. Our paper also includes a detailed bias analysis in credit assessment, addressing fairness concerns. Experimental results demonstrate that MASCA outperforms baseline approaches, highlighting the effectiveness of hierarchical LLM-based multi-agent systems in financial applications, particularly in credit scoring.

Problem

• Enhancing credit assessment using LLM-based multi-agent systems
• Addressing bias and fairness in credit scoring methods
• Optimizing decision-making with contrastive learning and game theory

Innovation

• LLM-driven multi-agent system for credit assessment
• Contrastive learning for risk and reward optimization
• Signaling game theory for hierarchical agent interactions