Interpretable LLMs for Credit Risk: A Systematic Review and Taxonomy

2025-06-04

AI Summary

This study addresses the interpretability bottleneck of large language models (LLMs) in credit risk assessment. Systematically applying the PRISMA framework, we identified and screened 60 empirical studies published between 2020 and 2025. We propose a novel four-dimensional taxonomy—covering model architecture, data modality, interpretability mechanisms, and application scenarios—and introduce the first domain-specific interpretability classification for LLMs in credit risk, distinguishing three innovation pathways: explanation mechanism design, prompt engineering optimization, and natural-language-based attribution. Through text analysis, cross-modal categorization, and interpretable pattern clustering, we construct the first LLM-based credit risk interpretability knowledge graph. This work fills a critical gap in standardized frameworks at the AI–finance intersection and provides both theoretical foundations and practical guidance for developing high-assurance, transparent credit decision-making systems.

Technology Category

• Machine Learning: Large Multimodal Models (LMMs)
• Natural Language Processing: Interpretability, Analysis, and Evaluation of NLP Models
• Computer Vision: Interpretability, Explainability, and Transparency

Application Category

• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation

Abstract

Large Language Models (LLM), which have developed in recent years, enable credit risk assessment through the analysis of financial texts such as analyst reports and corporate disclosures. This paper presents the first systematic review and taxonomy focusing on LLMbased approaches in credit risk estimation. We determined the basic model architectures by selecting 60 relevant papers published between 2020-2025 with the PRISMA research strategy. And we examined the data used for scenarios such as credit default prediction and risk analysis. Since the main focus of the paper is interpretability, we classify concepts such as explainability mechanisms, chain of thought prompts and natural language justifications for LLM-based credit models. The taxonomy organizes the literature under four main headings: model architectures, data types, explainability mechanisms and application areas. Based on this analysis, we highlight the main future trends and research gaps for LLM-based credit scoring systems. This paper aims to be a reference paper for artificial intelligence and financial researchers.

Problem

• Systematically reviews LLM-based credit risk estimation approaches
• Analyzes interpretability mechanisms for LLM credit models
• Identifies future trends and gaps in LLM credit scoring

Innovation

• Systematic review of LLM-based credit risk models
• Taxonomy covers model architectures and explainability
• Analyzes financial texts for risk assessment