AI-Driven Paradigms for Accelerating Battery Materials Discovery

In this talk, I presented our progress in applying machine learning and large language models to battery-material design and literature-driven catalyst screening.

Key content of the presentation includes:

• Integrating AI with battery science: Discussed how data-driven models can connect materials descriptors, electrochemical conditions, and battery performance.
• Interpretable machine learning: Presented data-analysis experience from hard-carbon materials for sodium-ion batteries and coordination-engineered catalysts for lithium–sulfur batteries.
• Emerging research directions: Discussed the potential of large language models and closed-loop workflows for scientific information extraction, catalyst evaluation, and materials discovery.