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

TL;DR

This paper introduces a novel approach that combines fine-tuned Large Language Models with XAI to improve human interpretability of ML models, demonstrated through a battery State-of-Health prediction case study.

Contribution

It proposes a reference architecture using fine-tuned LLMs as interactive chatbots for better ML interpretability, validated in a practical battery health prediction context.

Findings

Prototype enhances interpretability for less experienced users

Evaluation shows improved understanding of ML models

Demonstrates effectiveness in battery health prediction

Abstract

Across various sectors applications of eXplainableAI (XAI) gained momentum as the increasing black-boxedness of prevailing Machine Learning (ML) models became apparent. In parallel, Large Language Models (LLMs) significantly developed in their abilities to understand human language and complex patterns. By combining both, this paper presents a novel reference architecture for the interpretation of XAI through an interactive chatbot powered by a fine-tuned LLM. We instantiate the reference architecture in the context of State-of-Health (SoH) prediction for batteries and validate its design in multiple evaluation and demonstration rounds. The evaluation indicates that the implemented prototype enhances the human interpretability of ML, especially for users with less experience with XAI.