Liu Group

AI-driven discovery of molecules and materials

Our group develops AI and computational methods to understand, predict, and design molecules and materials. We are particularly interested in generative AI and machine learning for atomistic modeling, bridging methodological advances in AI with fundamental and practical challenges in chemistry and materials science.

Predictive AI

We develop machine learning models for predicting molecular and materials properties across multiple scales, from electronic structures and interatomic interactions to macroscopic properties.

Generative AI

We develop generative models that explore chemical and materials space and enable inverse design toward desired structures and properties.

Physics- and Chemistry-Aware AI

We incorporate physical principles and chemical constraints into AI models to improve their interpretability, generalizability, and ability to navigate chemically feasible spaces.

AI Infrastructure

We develop open datasets, benchmarks, evaluation metrics, software, and web-based platforms that make AI models more accessible, reproducible, and useful for scientific discovery.