Enhancing XAI Narratives through Multi-Narrative Refinement and Knowledge Distillation

2025-10-03

AI Summary

Existing counterfactual explanation techniques reveal deep learning decision logic but suffer from complex, low-readability formulations that hinder comprehension by non-expert users. To address this, we propose a novel framework integrating multi-round narrative refinement and knowledge distillation: a large language model (LLM) acts as a teacher to guide a lightweight small language model (SLM) in generating natural-language counterfactual explanations; we further design a dual-dimension automated evaluation metric assessing both counterfactual validity and linguistic readability. Our approach preserves the SLM’s computational efficiency while substantially improving explanation quality—narrowing the performance gap with LMs. Experiments across multiple XAI benchmarks demonstrate simultaneous enhancement in explanation readability, factual fidelity, and practical utility. The framework thus offers a viable pathway for deploying trustworthy AI in resource-constrained environments.

Technology Category

• Humans and AI: Explainable AI (XAI) for Human Understanding
• Natural Language Processing: (Large) Language Models
• Machine Learning: Large Multimodal Models (LMMs)

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation

Abstract

Explainable Artificial Intelligence has become a crucial area of research, aiming to demystify the decision-making processes of deep learning models. Among various explainability techniques, counterfactual explanations have been proven particularly promising, as they offer insights into model behavior by highlighting minimal changes that would alter a prediction. Despite their potential, these explanations are often complex and technical, making them difficult for non-experts to interpret. To address this challenge, we propose a novel pipeline that leverages Language Models, large and small, to compose narratives for counterfactual explanations. We employ knowledge distillation techniques along with a refining mechanism to enable Small Language Models to perform comparably to their larger counterparts while maintaining robust reasoning abilities. In addition, we introduce a simple but effective evaluation method to assess natural language narratives, designed to verify whether the models' responses are in line with the factual, counterfactual ground truth. As a result, our proposed pipeline enhances both the reasoning capabilities and practical performance of student models, making them more suitable for real-world use cases.

Problem

• Simplifying complex counterfactual explanations for non-expert understanding
• Enhancing small language models' narrative generation through knowledge distillation
• Evaluating natural language explanations against factual counterfactual truths

Innovation

• Leveraging Language Models for counterfactual explanation narratives
• Using knowledge distillation to enhance Small Language Models
• Introducing evaluation method for natural language narrative alignment