Agentic CBR in action: empowering loan approvals through interactive, counterfactual explanations.

Salimi, Pedram; Wiratunga, Nirmalie; Corsar, David

Abstract

Large Language Models (LLMs) have demonstrated impressive conversational capabilities, yet their susceptibility to hallucinations and inconsistent recommendations poses significant risks in high-stakes domains such as finance. This paper presents an interactive chatbot for loan application guidance that leverages a case-based reasoning (CBR) approach to generate actionable counterfactual explanations within an agentic framework. Our system employs a supervisor agent, built using the LangGraph framework, to orchestrate four specialised agents: a classifier agent that provides an initial loan prediction, a causally-aware counterfactual explanation agent that proposes minimal yet feasible modifications to reverse an unfavourable decision, a Feature Actionability Taxonomy (FAT) agent that updates user-specific immutability constraints based on feedback, and a template-based natural language generation (NLG) agent that transforms counterfactual suggestions into clear, user-friendly explanations. A key strength of our design is the automated feedback loop: when users indicate that certain suggestions are unworkable, the FAT agent revises the constraints and instructs the counterfactual generation agent to produce a refined explanation. We detail the system architecture and workflow and outline an experimental plan that compares our full agentic chatbot to ablated variants and a LLM-Only Baseline. And finally we outline a planned user study to evaluate how controlled reasoning affects trust in high-stakes lending.

Citation

SALIMI, P., WIRATUNGA, N. and CORSAR, D. 2025. Agentic CBR in action: empowering loan approvals through interactive, counterfactual explanations. In Martin, K. and Ye, X. (eds.) ICCBR-WS 2025: joint proceedings of the workshops and doctoral consortium at the 33rd International conference on case-based reasoning (ICCBR-WS 2025) co-located with the 33rd International conference on case-based reasoning (ICCBR 2025), 30 June 2025, Biarritz, France. CEUR workshop proceedings, 3993. Aachen: CEUR-WS [online], pages 27-42. Available from: https://ceur-ws.org/Vol-3993/paper3.pdf

• Presentation Conference Type: Conference Paper (published)
• Conference Name: 33rd International conference on case-based reasoning workshops and doctoral consortium (ICCBR-WS 2025) co-located with the 33rd International conference on case-based reasoning (ICCBR 2025)
• Start Date: Jun 30, 2025
• Acceptance Date: Apr 6, 2025
• Online Publication Date: Jun 12, 2025
• Publication Date: Jul 8, 2025
• Deposit Date: Jul 31, 2025
• Publicly Available Date: Jul 31, 2025
• Publisher: CEUR-WS
• Peer Reviewed: Peer Reviewed
• Pages: 27-42
• Series Title: CEUR-workshop proceedings
• Series Number: 3993
• Series ISSN: 1613-0073
• Book Title: ICCBR-WS 2025
• Keywords: Conversational AI; Counterfactual explanations; Agentic workflow; CBR; Large language models (LLM); Hallucinations
• Public URL: https://rgu-repository.worktribe.com/output/2941443
• Publisher URL: https://ceur-ws.org/Vol-3993/