Sign in

Davide Tisi

@dtisi.bsky.social
80 followers 137 following 6 posts

Condensed matter physicist working with machine learning, PhD at SISSA and currently a post-doc in Ecole Polytechnique Fédérale de Lausanne at @labcosmo.bsky.social

PostsRepliesMedia
Reposted by Davide Tisi
APS journals @apsjournals.aps.org · 30/08/2026
Using a universal machine-learning potential, researchers mapped how halogen and metal substitution impact the structure, phase stability, and lithium ion conductivity of halide solid electrolyte compounds. Read the Physical Review Materials paper: go.aps.org/4c85mEn
On the left is a graph labeled phase stability with Cl content (%) on the x axis and energy (meV/atom) on the y axis. It depicts the formation energy for the P¯3m1 and C2/m structures as a function of Cl content, computed with both the PET-MAD potential and its fine-tuned version. The dotted line represents the ideal solution entropy. On the right is a graph labeled Li+ conductivity, with content Cl (%) on the x axis and σ (S/m) on the y axis. It shows the ionic conductivity of Li3YBr6(1−x)Cl6x computed via the PET-MAD universal potential, shown in blue, and a PET model fine-tuned, shown in pink. The shaded area indicates the statistical uncertainty for the pure Li3YBr6 and Li3YCl6 phases; for the mixed composition, it indicates the semi-dispersion among the 4 from the different simulations selected during the MC procedure. Results are compared with previous experimental results, showing Z. Liu et al in green and E. van der Maas et al. in brown.
021
Reposted by Davide Tisi
COSMO Lab @labcosmo.bsky.social · 09/12/2025
No day goes by without a new universal #ML potential. But how different they really are? Sanggyu and Sofiia tried to give a quantitative answer by comparing the reconstruction errors between their latent-space features. If you are curious, check out the #preprint arxiv.org/html/2512.05...
Features reconstruction errors between the latent spaces of several universal MLIPs
0113
Reposted by Davide Tisi
COSMO Lab @labcosmo.bsky.social · 10/12/2025
Congrats to 🧑‍🚀 Sergey Pozdnyakov who received a distinction (best 8% of theses at @materials-epfl.bsky.social) for his PhD thesis "Advancing understanding and practical performance of machine learning interatomic potentials". Поїхали 🚀! infoscience.epfl.ch/entities/pub...
0102
Reposted by Davide Tisi
COSMO Lab @labcosmo.bsky.social · 28/11/2025
📢 PET-MAD is here! 📢 It has been for a while for those who read the #arXiv, but now you get it preciously 💸 typeset by @natcomms.nature.com Take home: unconstrained architecture + good train set choices give you fast, accurate and stable universal MLIP that just works™️ www.nature.com/articles/s41...
nature.com
PET-MAD as a lightweight universal interatomic potential for advanced materials modeling - Nature Communications
PET-MAD is a fast and lightweight universal machine-learning potential, trained on a small but diverse dataset, that delivers near-quantum accuracy in atomistic simulations for both organic and inorga...
0156
Reposted by Davide Tisi
Materials Science & Engineering @EPFL @materials-epfl.bsky.social · 19/09/2025
👏 Congratulations to Prof. @micheleceriotti.bsky.social (@labcosmo.bsky.social) and Prof. Roland Logé (LMTM) on their promotion to Full Professor of Materials Science in the School of Engineering! More info: actu.epfl.ch/news/appoint...
231
Reposted by Davide Tisi
COSMO Lab @labcosmo.bsky.social · 27/08/2025
📢 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
194
Davide Tisi @dtisi.bsky.social · 22/08/2025
Just before my last week in @labcosmo.bsky.social. Our metatensor and metatomic paper is out! A collection of the hard work we’ve done at @labcosmo.bsky.social to make atomistic machine learning easier to use for experts and not alike.
010
Reposted by Davide Tisi
COSMO Lab @labcosmo.bsky.social · 27/05/2025
📢 Running molecular dynamics with time steps up to 64fs for any atomistic system, from Al(110) to Ala2? Thanks to 🧑‍🚀 Filippo Bigi and Sanggyu Chong, with some help from Agustinus Kristiadis, this is not as crazy as it sounds. Let us briefly introduce FlashMD⚡ arxiv.org/html/2505.19...
Scheme of the GNN architecture of the FlashMD method.
13712
Reposted by Davide Tisi
COSMO Lab @labcosmo.bsky.social · 07/05/2025
The PET-MAD universal forcefield mingled with the dark side, and got twice as fast 🚀. Read on, or head to the 🧑‍🍳📖 atomistic-cookbook.org/examples/pet..., if you are curious of what this is all about. #atomistic-cookbook #compchem #machinelearning #mlip🧵
Plot of the potential and total energy trends along a MD trajectory, showing drift for non-conservative forces, and how that is fixed using multiple time stepping.
1103
Davide Tisi @dtisi.bsky.social · 21/04/2025
Last week, we continued @labcosmo.bsky.social's study of solid-state electrolytes, focusing on the properties of the LPS surfaces arxiv.org/abs/2504.11553. Many thanks to my office mate Hanna Tuerk, who did a lot of the work
arxiv.org
Reconstructions and Dynamics of $β$-Lithium Thiophosphate Surfaces
Lithium thiophosphate (LPS) is a promising solid electrolyte for next-generation lithium-ion batteries due to its superior energy storage, high ionic conductivity, and low-flammability components. Des...
131
Reposted by Davide Tisi
COSMO Lab @labcosmo.bsky.social · 19/03/2025
Find (many) more benchmarks and tests in the preprint arxiv.org/html/2503.14..., try PET-MAD for yourself, and let us know if you manage to break it - we're already working towards PET-MAD-2 😅 #compchem @nccr-marvel.bsky.social @materials-epfl.bsky.social @erc.europa.eu @cscsch.bsky.social
arxiv.org
PET-MAD, a universal interatomic potential for advanced materials modeling
081
Reposted by Davide Tisi
Sven Rogge @svenrogge.bsky.social · 09/04/2025
@labcosmo.bsky.social continuing its strong performance with Davide Tisi's talk on transport mechanisms in solid-state electrolytes!
042
Reposted by Davide Tisi
COSMO Lab @labcosmo.bsky.social · 19/03/2025
📢 PET-MAD has just landed! 📢 What if I told you that you can match & improve the accuracy of other "universal" #machinelearning potentials training on fewer than 100k atomic structures? And be *faster* with an unconstrained architecture that is conservative with tiny symmetry breaking? Sounds like 🧑‍🚀
Polar plot showing the errors of several machine-learning potential of different test sets. Smaller is better here!Plots showing the evaluation time per atom for several machine-learning potentials as a function of the number of atoms in a simulation. Smaller is better
1289