Cover Feature: Adaptive Pruning for Increased Robustness and Reduced Computational Overhead in Gaussian Process Accelerated Saddle Point Searches (ChemPhysChem 4/2026)

The Cover Feature illustrates the use of local surrogate energy surfaces for saddle searches, with rabbits and carrots indicating the progress of the algorithm and the samples taken. Pink contours show the fast Gaussian process surrogate model, while the violet surface signifies the computationally expensive true potential energy surface. Thehighernumberof samples on the surrogate translates to computational efficiency in the number of samples and total time. The inset highlights the second major innovation of the optimal transport Gaussian process, described in the Research Article by R. Goswami and H. Jónsson (DOI: 10.1002/cphc.202500730): how the distance between two configurations is computed in terms of solving a separate Earth mover's distance for each element.