Lightnews — Scholar-powered news

Reposted by COSMO Lab

Michele Ceriotti @micheleceriotti.bsky.social · 3d

Looks like @ox.ac.uk forbids their researchers to do any kind of literature search, though it seems that thankfully they can still submit to the arxiv arxiv.org/abs/2510.00027 🤷

Learning Inter-Atomic Potentials without Explicit Equivariance

Accurate and scalable machine-learned inter-atomic potentials (MLIPs) are essential for molecular simulations ranging from drug discovery to new material design. Current state-of-the-art models enforc...

arxiv.org

2 6

COSMO Lab @labcosmo.bsky.social · 16d

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?

COSMO Lab @labcosmo.bsky.social · 16d

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 ...

1

COSMO Lab @labcosmo.bsky.social · 16d

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.

1

COSMO Lab @labcosmo.bsky.social · 16d

📝 We have been told (& been telling) that ML potentials are linked quite directly to the expansion of the atomic energy into pairs, triples, and so on. But is this actually true 🤔? Go read the latest from the 🧑‍🚀 team (w/QM help from Joonho's team at Harvard) to find out more arxiv.org/html/2509.14...

Resolving the Body-Order Paradox of Machine Learning Interatomic Potentials

arxiv.org

1 1 4

COSMO Lab @labcosmo.bsky.social · 20d

Bragging time - ⚡ FlashMD⚡ was accepted as a spotlight paper at #NeurIPS25. if you still haven't checked it out, it's already on the #arxiv arxiv.org/abs/2505.19350, the code is at flashmd.org and the 🧑‍🍳📖 is here atomistic-cookbook.org/examples/fla.... Congrats to Filippo, Sanggyu and Augustinus!

GitHub - lab-cosmo/flashmd: A universal ML model to predict molecular dynamics trajectories with long time steps

A universal ML model to predict molecular dynamics trajectories with long time steps - lab-cosmo/flashmd

flashmd.org

2

Reposted by COSMO Lab

Christoph Dellago @chhdellago.bsky.social · 28d

Michele Parrinello giving the ICTP Colloquium (he speaks about catalysis) as part of the conference celebrating his 80th birthday. Amazing creativity throughout a long career!

1 3 15

Reposted by COSMO Lab

Jennie Martin @jennieemartin.bsky.social · Sep 3

I'm very pleased to say my first preprint, with @graemeday.bsky.social and @micheleceriotti.bsky.social is now online!

This is the main work of my PhD, adapting a similarity kernel to be more suited for exploring molecular CSP landscapes

#compchemsky #chemsky #compchem

doi.org/10.26434/che...

We adapted an existing approach to identifying stabilisable crystal structures from prediction sets - the Generalised Convex Hull (GCH) - to improve its application to molecular crystal structures. Th...

doi.org

2 6

COSMO Lab @labcosmo.bsky.social · Aug 28

Anticipating 🧑‍🚀 Wei Bin's talk at #psik2025 (noon@roomA), 📢 a new #preprint using PET and the MAD dataset to train a universal #ml model for the density of states, giving band gaps for solids, clusters, surfaces and molecules with MAE ~200meV. Go to the talk, or check out arxiv.org/html/2508.17...!

error plots for the PET-MAD-DOS model on different datasets

2 3

COSMO Lab @labcosmo.bsky.social · Aug 27

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

2

COSMO Lab @labcosmo.bsky.social · Aug 27

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!

Reaction energies of pristine and reconstructed surfaces with water

1 1

COSMO Lab @labcosmo.bsky.social · Aug 27

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

Surface energy diagram of LPS before and after reconstruction

Wulff shape of LPS particles based on the computed surface energies

1 1

COSMO Lab @labcosmo.bsky.social · Aug 27

📢 Now out on @physrevx.bsky.social energy, journals.aps.org/prxenergy/ab... from 🧑‍🚀 @dtisi.bsky.social and Hanna Türk, our #PET -powered study of the dynamic reconstruction of LPS surfaces, and how it affects their structure, stability and reactivity.

A cartoon explaining how mild finite-temperature conditions induce disorder and dynamical reconstruction on the surfaces of lithium thiophosphates

1 4 8

COSMO Lab @labcosmo.bsky.social · Aug 26

If you are at the #psik2025 and want to know more about the #metatensor ecosystem, don't miss @luthaf.bsky.social talk tomorrow morning 9:45 in room 1