🚀 After two+ years of intense research, we’re thrilled to introduce Skala — a scalable deep learning density functional that hits chemical accuracy on atomization energies and matches hybrid-level accuracy on main group chemistry — all at the cost of semi-local DFT. ⚛️🔥🧪🧬

🚀 After two years of intense research, we’re thrilled to introduce Skala — a scalable DL density functional that hits chemical accuracy on atomization energies and matches hybrid-level performance on main group chemistry — all at the cost of a semi-local functional. ⚛️🔥🧪⚗️🧬

Really proud to have worked on this during my internship with such an incredible team! DFT functionals could become one of the most positively impactful applications of scientific ML in the next years 🚀

This is excellent!

Congrats John Gardner Gregor Simm and other contributors! Great work studying multi-fidelity training of ML force fields and different pre-training and multitask learning schemes.

This is a game-changer! It's a privilege to collaborate with the incredible team at Microsoft Research #AI4Science, pushing the boundaries of #DFT to achieve unprecedented accuracy with #DeepLearning. Blog: msft.it/6011SQwKX Arxiv: arxiv.org/abs/2506.14665 #CompChem #CCSD(T)

Finally had a chance to read this and I have to say it's pretty impressive. Huge accuracy improvements without any exact exchange, let alone local EXX like DM21. The accuracy on MB43-16 is insane. That said, it probably fails for transition metals...

MSR AI Frontiers has exciting new openings. It's a growing team with opportunities for real world impactful research. Apply or reach out if you are interested!

Another exciting electronic structure paper from our team finally out. Check it out, deep QMC for strongly correlated systems!

🚨 Job alert

🎉 Great news: Our Machine Learning and Physical Sciences workshop at NeurIPS Conference will be back again this year! 🎉 Keep an eye out for updates on deadlines etc, we will be updating the website soon ml4physicalsciences.github.io #ML4PS2025 Machine Learning And The Physical Sciences

Today in the journal Science: BioEmu from Microsoft Research AI for Science. This generative deep learning method emulates protein equilibrium ensembles – key for understanding protein function at scale. msft.it/6010S7T8n

Congratulations to Danilo J. Rezende and Shakir Mohamed! This is how first learned about normalizing flows. Their paper on VAEs could/should have also gotten a nod.

🧵 Everyone is chasing new diffusion models—but what about the representations they model from? We introduce Discrete Latent Codes (DLCs): - Discrete representation for diffusion models - Uncond. gen. SOTA FID (1.59 on ImageNet) - Compositional generation - Integrates with LLM 🧱

Aurora is fully open! 🥳 The air pollution model 🌬️, the ocean wave model 🌊, and the TC tracker 🌀 are now available. And that's not all: all model weights (pretrained and fine-tuned) are now released under the MIT license. 😎 GitHub: github.com/microsoft/auro… #AIforGood

Want to join our efforts Microsoft Research AI for Science to push the frontier of AI for materials? We are the team behind MatterGen & MatterSim and we have 2 job openings! Each can be in Amsterdam, NL, Berlin, DE, or Cambridge, UK. It is a rare opportunity to join a highly talented,

Exciting work by our team, representing a real step forward in understanding the protein dynamics that power biological function.

Exciting news! We're thrilled to announce the appointment of Professor Hugo Larochelle as Mila's new Scientific Director! A deep learning pioneer and former head of Google's AI lab in Montreal, Hugo's leadership will be pivotal in advancing AI for the benefit of all. Read the