marcel ⊙

marceldotsci.bsky.social

18 September 2026 10:00

✨ new work: "Truncated automatic sparse differentiation for MLIPs" with @adrianhill.de & @micheleceriotti.bsky.social has arrived on the arXiv. In a 🥥: we exploit that MLIP Hessians are sparse and decay with distance with a "truncated" variant of automatic sparse differentiation. See figure. 👇

COSMO Lab

labcosmo.bsky.social

03 March 2026 22:11

So let us show you just how *universal* #PET-MAD-1.5 can be. This is a movie of a parallel tempering simulation, with replicas from 300K to 3000K, of what we call a "Mendeleev cluster" - one atom each of every element from 1 to 102.

COSMO Lab

labcosmo.bsky.social

18 February 2026 18:41

📢 New #preprint is out! Investigating the many flavors of last-layer #UQ, Moritz and 🧑‍🚀Matthias propose a practitioners' guide on "how to train a shallow ensemble". TL;DR? for good calibration use NLL, include force, and optimize the backbone, fine-tuning for speed! 📃🔗➡️ arxiv.org/html/2602.15...

COSMO Lab

labcosmo.bsky.social

08 February 2026 14:14

If you want to learn about materials modeling, from DFT to MD, well marinated in a spicy ML sauce, don't miss out the @ictp.bsky.social / @nccr-marvel.bsky.social college. Details and application instructions here indico.ictp.it/event/11146/. See you in Miramare!

COSMO Lab

labcosmo.bsky.social

14 January 2026 11:40

Not going to make a big deal out of a benchmark table, but PET just got the top spot on matbench-discovery.materialsproject.org. And don't be fooled by the huge parameters count, it's faster and can handle larger structures than eSEN-30M 🚀. Kudos to 🧑‍🚀 Filippo, Arslan and Paolo!