CreditR1 delivers calibrated PDs with evidence-grounded reasoning that supports internal model validation and human review that supports transferability beyond the Chinese A-share market remains an open empirical question.
Abstract
Probability-of-default (PD) estimation under the Basel and IFRS 9 frameworks requires probabilities that are both discriminative and well-calibrated. Large language models applied to credit assessment via prompting or supervised fine-tuning (SFT) yield poorly calibrated probabilities, while reinforcement learning with binary correctness rewards is structurally unsuitable for probability prediction under extreme class imbalance. We propose CreditR1, a three-stage framework: an SFT cold start on evidence-filtered reasoning chains; Group Relative Policy Optimization, guided by a composite verifiable reward combining Brier-score calibration—a strictly proper scoring rule—pairwise ranking, evidence anchoring, and format compliance; and an anti-contamination evaluation protocol. On Chinese A-share corporate credit data, CreditR1 matches gradient-boosted baselines in discrimination (AUC: 0.883±0.004 vs. 0.891 for XGBoost) while reducing expected calibration error by 24.2% versus isotonic-calibrated XGBoost (ECE: 0.047 vs. 0.062) and by 47.2% versus uncalibrated XGBoost (0.089). Because the test set contains only 119 default events, all comparisons carry firm-level bootstrap confidence intervals; the calibration advantage remains significant against every baseline after Holm–Bonferroni correction, including Platt, beta, and Bayesian-binning recalibrations. Ablations confirm each reward component is necessary. CreditR1 delivers calibrated PDs with evidence-grounded reasoning that supports internal model validation and human review; transferability beyond the Chinese A-share market remains an open empirical question.
This work examines whether Large Language Models (LLMs) can serve as explanation layers that translate post-hoc explanation artefacts into stakeholder-appropriate risk narratives and discusses implications for the governance of risk models, including deployment considerations and the value of domain-aligned LLMs in regulated credit settings.
Sahab Zandi, Noah Kostesku, Christophe Mues et al.· 0 citations
Machine learning is widely used to improve predictive accuracy in complex domains like credit scoring, but many models (e.g., deep neural networks) remain opaque. This lack of interpretability is problematic in regulated domains (banking, finance) where transparency is required. Rule-based learning methods, such as Learning Classifier Systems (LCS), offer a trade-off between accuracy and explainability. We introduce a novel Ranked Attribute Selection with Midpoint Filtering (RASF) framework that extends LCS (EXTRACS) to enhance feature selection and rule validation for credit approval. RASF first ranks features by mutual information, then employs rank-guided randomized selection to diversify rule conditions, and finally filters new rules by midpoint-based Euclidean distance to the current instance. We evaluate RASF-enhanced LCS on public loan approval datasets, comparing against a baseline LCS and logistic regression. Results show that RASF improves predictive accuracy by about 3.6 – 4.64% over the base LCS, while producing an inherently interpretable rule set. By bridging accuracy and transparency, RASF-LCS supports explainable AI in credit scoring.
M. Ahamed, Abubakar Siddique, Trung Nguyen et al.· Proceedings of the Genetic a...· 0 citations
In the ever-evolving financial landscape, dynamic credit limit adjustment plays a critical role in optimizing customer experience and risk management. Traditional methods often rely on static, rule-based systems that lack adaptability and transparency. This paper proposes an Explainable Deep Reinforcement Learning (XDRL) framework to automate and personalize credit limit adjustments based on customer behavior, financial data, and macroeconomic indicators. Our model learns optimal credit limit strategies that balance risk, customer satisfaction, and profitability. To ensure transparency and regulatory compliance, we integrate explainability modules—such as SHAP values and attention mechanisms—into the DRL pipeline. We evaluate the system on real-world or simulated credit data, demonstrating improvements in credit utilization, default prediction, and interpretability. This work paves the way for safer, more accountable AI-driven decision-making in the credit industry.
B. K, Meena Krishnan· International Journal of Int...· 0 citations
This work proposes a tenure-stratified hybrid framework that couples an online weight-of-evidence logistic regression (WOE-LR) scorecard with an offline self-attention stacked ensemble whose calibrated PD is quantile-binned, WOE-encoded, and re-injected into the online scorecard as a single auditable predictor.
Gulnaz Zakariya, A. Moldagulova, Nor’ashikin Ali· Big Data and Cognitive Compu...· 0 citations
Standard credit-risk scorecards rely on linear ratio thresholds that break down when feature interactions are nonlinear and observations carry temporal dependencies. Qualitative signals embedded in corporate disclosures—tone shifts, forward-looking hedges, and sector-specific terminology—remain largely ignored by numeric-only models, even though such signals often precede ratio deterioration. This paper introduces a tri-modal deep learning framework that jointly trains three complementary branches: a Convolutional Neural Network (CNN) for cross-sectional ratio-pattern detection, a Long Short-Term Memory (LSTM) network for multi-quarter trend modelling, and a Natural Language Processing (NLP) branch for disclosure-text encoding. Prior to deep-model training, LASSO regularisation removes collinear financial indicators and SMOTE oversampling corrects the severe class imbalance characteristic of distress datasets. A feature-concatenation fusion layer integrates all three branch outputs; the resulting vector feeds a sigmoid classifier that produces a calibrated distress probability. Benchmarked against five baselines on four financial datasets, the model reaches 94.8% accuracy and 91.3% minority-class recall, with a 4.1-point F1 advantage over the strongest single-modality competitor.
Paiinti Meenakshi, Muthaluru Bhuvaneshwari, Jalla Ganesh et al.· International Conference Com...· 0 citations