Aug 2026
Robust generative transition-state models for unseen chemistry.
Self-supervised pretraining substantially improves TS prediction for previously unseen systems, lowering the median root-mean-square deviation of TS geometries on Transition1x-TMC reactions and reducing fine-tuning data requirements, enabling reliable performance even in low-data regimes.
Samir Darouich, Jacob W. Toney, Weiliang Luo et al.
· Nature Computational Science · 0 citations