Accurate Ensembles, Fragile Narratives: Multi-Scale Stacking and a Fidelity Audit of LLM-Generated Explanations for Credit Risk
Gregorius Reynaldi Pratama, Kuo-Kun Tseng
cs.LG, cs.CL
Submitted: 2026-08-08
Updated: 2026-08-11
Comments: 13 pages, 9 figures, 5 tables
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Terminology
Sources
- Can formal argumentative reasoning enhance LLMs performances?
- XAI for All: Can Large Language Models Simplify Explainable AI?
- In-Context Explainers: Harnessing LLMs for Explaining Black Box Models
- From XAI to Stories: A Factorial Study of LLM-Generated Explanation Quality
- Could Large Language Models work as Post-hoc Explainability Tools in Credit Risk Models?
- A Two-Stage LLM Framework for Accessible and Verified XAI Explanations
- Interpreting LLMs as Credit Risk Classifiers: Do Their Feature Explanations Align with Classical ML?
- On the Robustness of Interpretability Methods
- Searching for Activation Functions
Related papers
- Polynomial-Augmented Neural Networks (PANNs) with Weak Orthogonality Constraints for Enhanced Function and PDE Approximation
- AIRL-S: Unifying Reinforcement Learning and Search-Based Test-Time Scaling via Adversarial Inverse Reinforcement Learning
- Transformers as Bayesian In-Context Experimenters: Smoothness-Adaptive Efficient ATE Estimation
- Convergence issues in Relational Concept Analysis based on AOC-posets
- Beliefs Beyond Posteriors: The Predictive Cost of ELBO Binding
- Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks