🔥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

The Geometric Algebra Transformer (GATr) is now boosted to high-energy physics! Great results on three tasks, including the first (afaik) Lorentz-invariant generative model. Amazing work by Jonas Spinner and Víctor

🚀We're excited to emerge from stealth and announce our $30M Seed financing round. A huge thank you to our incredible investor group including Hoxton Ventures, Basis Set, Lightspeed, Northzone, LocalGlobe, Touring Capital, Giant Ventures, FJ Labs, Zero Prime VC & Tiferes

I’m excited to share that I have joined the advisory board of CuspAI who have today come out of stealth mode and are using cutting edge AI to tackle one of the most urgent problems we as society face: climate change.

**Are language models (LMs) capable of self-recognition?** A growing number of apps are built on just a few frontier LMs- should we be worried about self-recognition? We use LM-generated “security questions” to find out! Our results: probably not, but for surprising reasons >🧵

Coming to devour your 3D anatomies: 🥼LaB-GATr🐊, geometric algebra transformers for large, biomedical surface and volume meshes. Will be presented MICCAI and Gram Workshop. Joint work with Baris Imre, Pim de Haan and Jelmer Wolterink. (code) github.com/sukjulian/lab-…

🐊 LaB-GATr is a transformer neural network designed for large-scale biomedical surface. It is inspired from GATr (pronounced gator 🐊), the Geometric Algebra Transformer (GATr) that cleverly represents inputs and states using projective geometric algebra. A much needed tutorial!

Very cool! Our recent work on "self-recognition in language models" is featured in IEEE Spectrum 🪞🤖❔ > Check out the article + paper/blog for more details: 📰 article: spectrum.ieee.org/self-aware-ai 🎓 paper: arxiv.org/abs/2407.06946 📝 blog: trdavidson.com/self-recogniti…

Today, I've joined CuspAI! With simulation and machine learning, we're going discover new materials for carbon capture. I am so excited to work with Max Welling, Chad Edwards and the rest of this team on this super important mission! Also: we're hiring!

Very cool that our paper on imitation learning with confounders got published! Congrats to Risto Vuorio

Join us in fighting the climate crisis with ML!

Coming to NeurIPS: learning the symmetries of dynamical systems with Noether's theorem! Congrats to amazing first author Tycho van der Ouderaa

CuspAI made it to the top 25 UK Startups to watch! bloomberg.com/features/2024-…

90% of researchers at big techs: We have multiple internship positions available in my team. Come and work with us on LLMs and AGI! Also 90% of researchers at big techs:

Another exciting day with Michael Albergo giving his perspective on generative model for physics, followed by Pim de Haan with exciting new work on CNFs for Gauge theories and Christopher Anders talking about a software package for flows in LFT! 👀🤩

We got some exciting new work out, led by the amazing Johann Brehmer. We find that equivariance matters at even scale! Not because of data-efficiency (augmentation solves that), but because of compute efficiency! Read the paper for the fine print.

Thrilled to announce that L-GATr is going to NeurIPS 2024! Plus, there is a new preprint with extended experiments and a more detailed explanation. Code: github.com/heidelberg-hep… Physics paper: arxiv.org/abs/2411.00446 CS paper: arxiv.org/abs/2405.14806 1/7

Today at NeurIPS, we’ll be presenting our Noether's Razor paper! 📜✨ 📅 Today Fri, Dec 13 ⏰ 11 a.m. – 2 p.m. PST 📍 East Exhibit Hall A-C, #4710 (ALL the way in the back I believe!) w/ Mark van der Wilk Pim de Haan Come say hi! 👋

Our paper got a prize :) Cheers to lead author Johann Brehmer, and fellow co-authors Sönke Behrends, and Taco Cohen. Our results hint that yes, also at large scale of data and compute, if your data has symmetries, you might be better off building these into your network.

Congratulations to Johann Brehmer Pim de Haan Taco Cohen ! Thanks for saving equivariance.

Please follow me on Pim den Hartog.bsky.social Long overdue to leave this sinking ship