Signal2026-07-21
arXiv

Correcting DFT formation energies towards experimental accuracy using foundational MLIPs and latent-feature delta-learning

Open primary sourceRead on Substack