uMLIPs-ft

Universal machine learning interatomic potentials (uMLIPs) represent a significant advancement in interatomic potential modeling, offering remarkable predictive accuracy across a wide range of chemical system. In this study, we evaluate two publicly available uMLIPs and investigate fine-tuning strategies to enhance their performance for catalytic reaction prediction. We systematically compared the fine-tuning and training from scratch in terms of predictive accuracy, data efficiency, and generalization capability.