Enhancing ML Model Interpretability: Leveraging Fine-Tuned Large Language Models for Better Understanding of AI

LLMs can make AI explanations more accessible: The paper demonstrates that fine-tuned large language models can translate complex machine learning explanations into more understandable formats through interactive chatbots, particularly benefiting users without technical expertise in explainable AI.

Black-box AI models are driving demand for better explanations: The research highlights that as machine learning models become increasingly difficult to understand ("black-boxedness"), there is growing momentum across sectors to develop explainable AI (XAI) solutions that help stakeholders understand how AI systems make decisions.

Interactive interfaces improve AI transparency: The prototype chatbot interface tested in the paper shows that interactive, conversational approaches to explaining AI decisions enhance human interpretability compared to traditional explanation methods, suggesting that how explanations are presented matters significantly.

Practical validation in real applications is important: The authors validated their approach specifically in battery State-of-Health prediction, demonstrating that responsible AI solutions should be tested in concrete, real-world applications rather than remaining theoretical—an important consideration for organizations implementing XAI.

Bridging the expertise gap supports responsible AI adoption: By making AI interpretability accessible to non-experts through LLM-powered tools, this work addresses a key barrier to responsible AI implementation, enabling broader organizational participation in understanding and governing AI systems.