I'm excited to host Peter Škrinjar to chat about the dataset/benchmarking techniques used to identify some of the memorization issues with the latest models like AlphaFold3, Chai-1, Protenix, and more and why it wasn't easy to spot until now. We'll leave lots of time for Q&A!

🚀 Antiviral Challenge Results! 🚀 We're excited to announce the conclusion of the ASAP Discovery x OpenADMET Antiviral Challenge! A huge thank you to everyone who participated, collaborated, and shared learnings, helping us collectively advance open science. 🎯 Challenge

1/ At Valence Labs, Recursion's AI research engine, we’re focused on advancing drug discovery outcomes through cutting-edge computational methods Today, we're excited to share our vision for building virtual cells, guided by the predict-explain-discover framework 🧵

It was really fun working on the app with the TxPert team at Valence! With the app, you can select a cell type, pick a perturbation, visualize a network of related genes, and select as many genes as you want to observe the transcriptional responses. Check it out 👇

Three weeks ago, we started building an AI game engine. But some models kept making things look... sloppy. So we turned finding the best one into a game. In three weeks, that game grew to 35K+ users across 135 countries. Introducing Design Arena, the fastest-growing

RNA is an untapped modality with tremendous amounts of potential. I'm excited to share more about the vision we're building towards Blank Bio! Learn more about our foundation models and the tools we're building to accelerate RNA-based therapeutics: blank.bio

A new AI research lab called Blank Bio launched today. They are building new foundation models for mRNA, and are one of the strongest technical teams I've worked with. I'm particularly excited about their new model results, and the wide set of application areas for it.

🎉 Huge congratulations to my PhD student, Phil Fradkin , on launching his new company building RNA foundation models—now backed by Y Combinator! 🚀 Excited to see how his work will push the boundaries of RNA biology and therapeutics. Proud mentor moment! 💪🧬

Will scaling current nucleotide foundation models lead to a GPT-3 moment for biology? Probably not. Although models like Evo2 are impressive in many ways, their pre-training objectives can struggle to outperform supervised learning. I dive into why at: quietflamingo.substack.com/p/scaling-is-d…