Extracting human interpretable structure-property relationships in chemistry using XAI and large language models
TL;DR: Combining XAI with LLMs to uncover structure-property relationships in chemistry
Abstract: Explainable Artificial Intelligence (XAI) is an emerging field in AI that aims to address the opaque nature of machine learning models. Furthermore, it has been shown that XAI can be used to extract input-output relationships, making them a useful tool in chemistry to understand structure-property relationships. However, one of the main limitations of XAI methods is that they are developed for technically oriented users. We propose the XpertAI framework that integrates XAI methods with large language models (LLMs) accessing scientific literature to generate accessible natural language explanations of raw chemical data automatically. We conducted 5 case studies to evaluate the performance of XpertAI. Our results show that XpertAI combines the strengths of LLMs and XAI tools in generating specific, scientific, and interpretable explanations.
Submission Track: Full Paper Track
Application Domain: Natural Language Processing
Survey Question 1: Explainable AI (XAI) is often not accessible to non-experts in the field. We demonstrate by combining XAI with large language models, we can uncover molecular structure-property relationships while making XAI more accessible. We present an easy to use web-app called XpertAI where users can upload data and extract natural language explanations
Survey Question 2: It has been shown previously black-box model first followed by XAI can be used to understand molecular structure-property relationships in Chemistry. However, deep learning let alone XAI are not easily accessible/interpretable to non-computational chemists. Our XpertAI workflow show that LLMs can address this limitation and generate human interpretable explanations from raw data.
Survey Question 3: We use SHAP and LIME in the current version of our XpertAI workflow.
Submission Number: 30
Loading