I joined @periodiclabs in May. We’re building AI scientists + autonomous labs to create breakthroughs you can hold. Longer take: https://t.co/g4kMSiwYBP More on Periodic:

Excited to release BoltzGen which brings SOTA folding performance to binder design! The best part of this project has been collaborating with many leading biologists who tested BoltzGen at an unprecedented scale, showing success on many novel targets and pushing its limits! 🧵..

To adapt to today's world, crank down discount factor and crank up learning rate

Every AI researcher should read two essays. One of them you already know. The other is "More is Different" by P. W. Anderson.

Introducing SPG: Sandwiched Policy Gradient for Masked Diffusion Language Models We propose a new policy gradient algorithm, SPG, for diffusion large language models. SPG improves the accuracy over the previous state-of-the-art RL methods by 3.6% in GSM8K, 2.6% in MATH500, 18.4%

It’s sad to see great researchers leaving FAIR, a research org that was once the best place to do AI research.

Excited to share SoftMimic -- a new approach for learning compliant humanoid policies that interact gently with the world.

@periodiclabs is following the natural progression of reward functions since 2022 from human-preference models (in RLHF), from math + code verifiable RMs (in reasoning models), to reward functions grounded in the physical world (in end-to-end optimization against experiment)

Tinker provides an abstraction layer that is the right one for post-training R&D -- it's the infrastructure I've always wanted. I'm excited to see what people build with it. "Civilization advances by extending the number of important operations which we can perform without

excited to share what we’ve been up to! at Periodic Labs, we’re creating AI scientists they think, design experiments, and optimize for results they also run experiments by moving robot arms in an autonomous lab it’s a fun, multi-turn task, with lots of open problems for AI

I am proud to have helped build @periodiclabs as a founding team member. Creating AI scientists and autonomous labs, RL-maxxing our way there, is the mission of a lifetime. Looking forward to be back in January to keep advancing the science x AI frontier. Incredible team, blindly

🚀 Thrilled to share that I’ve joined Periodic Labs in my dream role! We’re building end-to-end AI systems that act as scientists in the real world. Most people talk about “AI scientists” as LLMs that reason about data. We’re aiming higher...

Today, my boyfriend @LiamFedus launched his company, Periodic Labs. I feel so lucky to have witnessed it from day one—Liam is the hardest-working person I know and the most passionate about science. From the start, our home became a gathering place for some of the brightest

Oh btw huge thanks to @johnohallman @aryaman2020 @bayeslord @tszzl @willccbb @dwarkesh_sp @itsclivetime @dylan522p @LiamFedus @tamaybes @AnjneyMidha for reading drafts of this essay and providing insightful feedback! The full substack (<5 min read) is linked below.

Why haven’t LLMs done new science yet? One reason from @periodiclabs cofounder @ekindogus: science is iterative. LLMs haven’t been able to iterate experimentally yet. “ LLMs can be very smart, but if they're not iterating on science, they won't discover science. To be honest,

@LiamFedus @ekindogus @periodiclabs Congratulations Liam & Dogus! Using AI to accelerate real world scientific explorations is an important pursuit

https://t.co/1cRso2fOmc

*Highly* recommend Slater's article on Periodic Labs. It's a historical walk-through of computation in the sciences and an identification of current bottlenecks

@LiamFedus @ekindogus @periodiclabs Congrats on the launch! Doing things in the physical world is underrated by AI people.

@LiamFedus @ekindogus @periodiclabs We're not going to reason our way to Dyson spheres. Excited for you and all the best to the team!

Very excited (millions of eV) to be working with @LiamFedus and @ekindogus and the rest of the @periodiclabs team. In the long arc of scientific history, there are a few key moments when things really accelerated — and Periodic is working on the next one. Below are some