FSscore has now been published in Chemistry-Methods (Chemistry Europe)! 🚀 Check out the updated paper on personalized synthesizability scoring here: …mistry-europe.onlinelibrary.wiley.com/doi/10.1002/cm…

We’ve released an improved version of SynFlowNet! SynFlowNet can now handle large action spaces (up to 220k building blocks) and we experiment with training different GFlowNet backward policies. (1/3) arXiv: arxiv.org/abs/2405.01155 code: github.com/mirunacrt/synf…

Our group is recruiting a PhD student for a project on combining traditional expert features with deep learning for reaction prediction. Reposts and spreading the word to interested people in your network appreciated! jobs.ethz.ch/job/view/JOPG_…

Excited about the intersection of AI and molecular sciences? Join us at the 'AI & the Molecular World' track AMLD Intelligence Summit in Lausanne on Feb 13! The call for talks is now open: forms.gle/Q77Quw92KfsMUw…

We are hiring (resharing appreciated)! Given a few recent successful grant applications (I got my SNSF Starting Grant 🚀), we are extending the LIAC (LIAC at EPFL) team and have multiple openings (PhD/postdoc) for 2025. Are you interested in #AI4Science and machine learning

I'm looking for an intern to join our drug discovery efforts at MSR AI for Science! If you're a student (not necessarily PhD) and have published work in AI4S, please apply ASAP; start date is flexible but we're currently targeting early spring. aka.ms/AI4ScienceInte…

1/ Starting material constrained synthesis planning is now possible using a general retrosynthesis algorithm *without* training a dedicated value network! Check-out our new paper, TangoStar. Preprint : arxiv.org/abs/2412.03424

Another chance to join our group! We are recruiting a PhD student in digital ligand engineering for nanocatalysis. Reposts and spreading the word to interested people in your network appreciated! jobs.ethz.ch/job/view/JOPG_…

🎉Our BoLudo paper is in J. Am. Chem. Soc.! Bayesian Optimization for nanocrystaL strUcture Design Optimization (jk, it's BOjana & LUDO😂) We show you can optimize ANYTHING when put on the right scale - even nanocrystals with surface energy! Also we discovered a new Cu shape! See how👇

LLMs are pretty bad at writing molecules, but quite good at analyzing mols and reactions! In our new work we use LLMs+search in chemical tasks, unlocking steerable synth. planning and mechanism prediction 🌟 1/ Theo Neukomm Daniel Armstrong Zlatko Jončev Philippe Schwaller (he/him) arxiv.org/abs/2503.08537

AdsMT is finally out in Nature Communications 🥳 It is designed for rapid prediction of global minimum adsorption energy (GMAE) from surface graphs and adsorbate descriptors. Thanks to Xu Huang , Philippe Schwaller (he/him) and NCCR Catalysis! Paper & code: nature.com/articles/s4146… LIAC at EPFL

🚨New preprint! ✨My precious✨ GOLLuM: GP-Optimized LLMs for Bayesian Optimization — the first of its kind! LLMs as deep kernels in GPs jointly optimized via marginal likelihood → implicit metric learning, calibrated uncertainty & superior sampling 🚀 📄 arxiv.org/abs/2504.06265

🔥 AutoAdsorbate 🔥 has been the workhorse of high throughput computation in heterogeneous catalysis modeling and ML BASF for the previous few years. 🧵 Now it is available open source! #MatSci #CompChem Preprint: doi.org/10.26434/chemr… Code: github.com/basf/autoadsor…

Check out Junwu Chen work on applying RL for goal-directed materials generation! Also check out Junwu's ablation experiments in the Appendix demonstrating the effect of the algorithm's components on *modulating* the optimization efficiency-sample diversity trade-off.