Introducing All-atom Diffusion Transformers — towards Foundation Models for generative chemistry, from my internship with the FAIR Chemistry team FAIR Chemistry AI at Meta There are a couple ML ideas which I think are new and exciting in here 👇

🚨 Check out DrugFlow, our new generative model for structure-based drug design. DrugFlow provides an atom-level confidence score for each designed molecule, and can adjust molecular size on the fly! Additional details in thread 🧵 #ICLR2025

#IBSevents: Friday 28/03 at 11am, seminar by Ilia Igashov EPFL, entitled 'Generative drug design : the next paradigm shift in therapeutic development?': urls.fr/RX2Gc- #DrugDesign

📢 Our new preprint is out on bioRxiv! We introduce RAG-ESM, a retrieval-augmented framework that improves pretrained protein language models like ESM2 by making them homology-aware with minimal additional training costs. 💻 github.com/Bitbol-Lab/rag… 🔗 doi.org/10.1101/2025.0… 1/7

I’ll be at #ICLR2025 in Singapore next week! I’ll present SynFlowNet - spotlight 🌟 poster no. 15 on Fri, 25/04 10am. Reach out if you want to chat!

Come to see our papers at #ICLR2025 in Singapore

Here are some recommendations for first timers in Singapore for #ICLR2025 ICLR 2026 that I’ve been sharing with friends! 🚕 How to get around the city?

Excited to be at #ICLR2025 and happy to chat about ML and drug discovery! 🇸🇬 Come find me at one of our posters or feel free to reach out directly! 1. DrugFlow: #9 Thu 24/04 10am 2. SynFlowNet: #15 Fri 25/04 10am 3. LDDM: GEMBio Workshop 27/04 4. LatentFrag: #AI4MAT 28/04

We'll be presenting the poster tomorrow (Thu 24/04) at 10am #ICLR2025. Please come and say hi :) iclr.cc/virtual/2025/p… Ilia Igashov Adrian Dobbelstein

Excited to present my first paper officially as a PhD student now as an ICLR Oral this week! Super fun work with the GenAIR team at NVIDIA. Talk: Fr 10:54 - 11:06 (Oral Session 3B, Garnet 213-215) Poster: Fr 15:00-17:30 (Hall 3 + Hall 2B #5) Come by the poster/reach out to chat

Excited to be at ICLR 2026 this week! I’ll be presenting our latest work at the MLGenX Workshop @ ICLR 2024 workshop on Sunday, details in the post below. (And yes, I’m on the job market, come say hi!)

Hello from Singapore 🇸🇬! Thrilled to be at #ICLR2025 presenting our work on fragment-based drug discovery 🧩. We go beyond virtual screening with a generative, structure-aware approach. 📃 openreview.net/forum?id=bZW1H… 🔗 github.com/rneeser/Latent… A thread 🧵👇

You'll find us at the SynFlowNet poster today between 10:00 and 12:30 in Hall 3 + 2B (poster #15) #ICLR2025

4 years ago, the Geometric Deep Learning proto-book hit the arXiv 📚 Today, we proudly release **Chapter 4** of the GDL Book! 📖 Wrapping up the foundations of GDL, exceeding the proto-book page count 🚀 Next up: deep-dives into the five Gs! #⃣🔁🕸️🌐🪢 Michael Bronstein @ICLR2025 🇸🇬 Taco Cohen

Explore the future of environmental forecasting with Aurora, an AI model from Microsoft Research. In this Abstracts episode, Megan Stanley & Wessel discuss how Aurora is redefining tropical cyclone and ocean wave forecasting. msft.it/6018Sdhn8

After a long hiatus, I've started blogging again! My first post was a difficult one to write, because I don't want to keep repeating what's already in papers. I tried to give some nuanced and (hopefully) fresh takes on equivariance and geometry in molecular modelling.

Excited to unveil Boltz-2, our new model capable not only of predicting structures but also binding affinities! Boltz-2 is the first AI model to approach the performance of FEP simulations while being more than 1000x faster! All open-sourced under MIT license! A thread… 🤗🚀

Super cool to see Boltz-2 paired with SynFlowNet! 🤩

Extremely excited to be sharing the output of my internship in Microsoft Research's #AIForScience team: "Understanding multi-fidelity training of machine-learned force-fields" 🤖🧪