However, this seems to damage the transferability of highly-preconditioned models such as MACE - less so for more expressive unconstrained models such as PET. Does this match your experience?

This doesn't matter much as most of the fragments that make up the body-order decomposition as deranged soups of highly-correlated electrons. Models with sufficient expressive power *can* learn if presented with the fragments ...

TL;DR: not really. ML potentials learn whatever they want, as long as it allows them good accuracy on the train set. We note in particular that MACE is strongly preconditioned to learn a fast-decaying body-order expansion, whether it decays fast or not.

With funding from a @snf-fns.ch Sinergia, the @nccr-marvel.bsky.social and @erc.europa.eu, and computing time from @cscsch.bsky.social !

The reconstructed surface contains different sites with different reactivity. Despite the higher stability, for some sites the disordered surface is *more* reactive with water, one of the main contaminants affecting the stability of LPS batteries. Useful to design better stabilization strategies!

Reconstructed surfaces become lower in energy, and the surface energy less orientation dependent - and so the Wulff shape of particles become more spherical.

Too many 👩‍🚀 and 👨‍🚀 are involved to list them all, but go check the contributors on GH. And this is a good time to thank the @nccr-marvel.bsky.social, @erc.europa.eu and the @snf-fns.ch which have given us the funding to dedicate to these #openscience efforts that don't make papers, but make science!

You can then load it and run from any software implementing an interface - lammps.org, ipi-code.org, ase-lib.org, plumed.org, theory.cm.utexas.edu/eon, radical-ai.github.io/torch-sim, chemiscope.org already do, and we're happy to help if you want an interface in your code.

TL;DR - this is a cross-platform, model-agnostic library to handle atomistic data (handling geometry and property derivatives such as forces and stresses) that lets you package your model into a portable torchscript file.

You can also try it out (if you like living on the edge, it's still in a branch github.com/lab-cosmo/fl...

We can get long-stride geometry-conserving integration by learning the Hamilton-Jacobi action. This fixes for good, doesn't just patch up, the instability of direct MD prediction, although it's not as fast. And work also for serious simulations, like glassy relaxation in deep supercooled GeTe!

Extremely sad news, I really hope IPAM somehow manages to carry on. If this is not seen as a successful program then I don't see what will be.