We show that our method successful generates low energy transitions for Alanine Dipeptide as well as the larger Polyproline and Chignolin proteins.

The OSSE NADIR and SWOT are pseudo-observations generated from the NEMO simulation.

Currently, two mainstream GNN-based methods have been developed for constructing force fields: group theory-based methods and direction-based methods.

A common approach to create more expressive GNNs is to change the message passing function of MPNNs. If a GNN is more expressive than MPNNs by adapting the message passing function, we call this non-standard message passing . Examples of this are message passing variants that operate on subgraphs [Frasca et al., 2022, Bevilacqua

Learning to simulate complex physics with graph networks.