Miao Liu

Formation energies from different exchange-correlation functionals have not been directly mergeable: in MatPES the same structures differ by 107 meV/atom on average between PBE and r2SCAN. Using more than 380,000 structurally paired PBE–r2SCAN entries, researchers trained a graph neural network to predict the energy residual between the two functionals and correct PBE formation energies onto the r2SCAN scale. With GPTFF as the backbone, the model reaches a mean absolute error of 14.3 meV/atom on the held-out test set, below the 18.2 meV/atom of CHGNet.

Formation energies from different exchange-correlation functionals have not been directly mergeable: in MatPES the same structures differ by 107 meV/atom on average between PBE and r2SCAN. Using more than 380,000 structurally paired PBE–r2SCAN entries, researchers trained a graph neural network to predict the energy residual between the two functionals and correct PBE formation energies onto the r2SCAN scale. With GPTFF as the backbone, the model reaches a mean absolute error of 14.3 meV/atom on the held-out test set, below the 18.2 meV/atom of CHGNet.