🔥The 9th Round of Easy Loan, Earn $40 Reward is in progress❗️ ⏰ Promotion Period: January 15th - Feburary 15th, 2025 👉 Register now and check more details at gate.io/campaigns/358

Our paper on computational design of chemically induced protein interactions is out in nature. Big thanks to all co-authors, especially Anthony Marchand, Stephen Buckley and Bruno Correia nature.com/articles/s4158…

Talking about simulations/structure prediction at UCL. And the quest to share the little I know and get more students into ML for Bio continues. Thank you CS Society team for the invitation!

NEW: Oxford University has been ranked first in the world in the Times Higher Education Subject Rankings for Medicine and Computer Science. Oxford leads in Medicine for the 14th consecutive year and in Computer Science for the 7th. More info ⬇️

As you can probably tell from our work, we're now all in on diffusion and flow matching models for protein/peptide sequence generation! 🌟 These models are just way better for tightly controlling generation (vs. say autoregressive models, which we've found are poorly set up for

Diffusion models leverage a variety of samplers. Deterministic methods like DDIM produce orderly paths. In contrast, stochastic samplers like DDPM produce chaotic trajectories. Despite their differences, both methods draw valid samples from the underlying distribution.

In the physical world, almost all information is transmitted through traveling waves -- why should it be any different in your neural network? Super excited to share recent work with the brilliant Mozes Jacobs: "Traveling Waves Integrate Spatial Information Through Time" 1/14

New paper (and #ICLR2025 Oral :)): ProtComposer: Compositional Protein Structure Generation with 3D Ellipsoids arxiv.org/abs/2503.05025 Condition on your 3D layout (of ellipsoids) to generate proteins like this or to get better designability/diversity/novelty tradeoffs. 1/6

Don't forget to join us this Friday with Karsten Kreis and Tomas Geffner to hear more about flow-based protein structure generation. 🗓️ Friday, Mar 14 ⏲️ 5PM CET / 12PM EDT / 9AM PDT Registration: genaiindrugdiscovery.com x.com/vant_ai/status…

Flow matching has gained popularity recently which is better, diffusion or flow matching? They are formally equivalent Our purpose is to help practitioners understand and use these frameworks interchangeably -- **regardless of what it’s called**

Extremely excited to give a glimpse of what we've been working on over the past 2 years: unifying structure-prediction and de-novo generation. Neo-1 is an all-atom, fully generative model

Heading to 🇸🇬 for #ICLR2025! Amazing work from our NVIDIA NVIDIA Healthcare teams the whole week, plus a great opportunity to visit our local colleagues. Come say hi!

I'll be at #ICLR2025 next week presenting a host of new work on generative models in #AI4Science. Please do reach out (DM's or otherwise) if you'd like to catch me at the conference. Main Conference: #1: The Superposition of Diffusion Models Using the Itô Density Estimator

🥇 The best paper award goes to RiboGen, authored by Dana Rubin, Allan Dos Santos Costa, Joseph Jacobson, and Manvitha Ponnapati! Thanks to NVIDIA, Bruno Trentini, and Daniel Burkhardt, for this great paper they received an RTX A6000 - 48GB GPU🔥

🥰: 1959, 1979, 1985, 1991, 1993, 2012 🤮: 1995, 2001 🤬: 1972 (((?))): 1958

Years of corporate social media training stopped me from becoming the greatest comedian and political comentator ever Or the internet’s ultimate shitposter. We’ll never know.

This is a great guide on Flow Matching from Meta featuring some incredibly intuitive visualizations. I found it to be a very accessible introduction to some of the theory behind flow based generative models. Link 👇

I don't know where the answer is But I'm gonna find it Eventually

Double your OpenMM MD simulation throughput with our multi-processing service in a single line of code. Learn how ➡️ nvda.ws/4mDxumi

Are you Chaotic, Neutral or Lawful?

🚨 ICML 2025 Paper 🚨 "On Measuring Long-Range Interactions in Graph Neural Networks" We formalize the long-range problem in GNNs: 💡Derive a principled range measure 🔧 Tools to assess models & benchmarks 🔬Critically assess LRGB 🧵 Thread below 👇 #ICML2025