Our Lyapunov paper in the New Scientist. Thanks Chris Stokel-Walker !

Transformers for discrete optimisation problems 1- Train a model on candidate solutions 2- Use the model to generate more candidates 3- Improve the solutions with local search 4- Use the best candidates to fine tune the model 5- Iterate

How do transformers learn arithmetic tasks, such as GCD and modular sums and products? My talk in Collège de France on November 4th (in French, but the English subtitles are quite good). Thank you Timothy Gowers @wtgowers for inviting me to your seminar! youtube.com/watch?v=e0jUi8…

I'm thrilled to announce that François Charton (François Charton AI at Meta) will be kicking off our new AI for Science seminar series next Wednesday. He is at the forefront of using AI for mathematics, cryptography, and theoretical physics. datascience@uw dsi.wisc.edu/2024/11/12/ai-…

Thank you for having me Kyle Cranmer and Gary Shiu Research featured in the talk: discovering Lyapunov functions (9:55), PatternBoost: generative models in combinatorics (31:20), Scattering amplitudes (44:10), Arithmetic, repetition, and a few unpublished results (46:40)

The code for our paper: Global Lyapunov functions: a long-standing open problem in mathematics, with symbolic transformers, with Alberto Alfarano and Amaury Hayat is available at github.com/facebookresear… We will be in NeurIPS: come see us at the poster session next Thursday at 5PM

One epoch is not all you need! Our paper, Emergent properties with repeated examples, with Julia Kempe, won the NeurIPS24 Debunking Challenge, organized by the Science for Deep Learning workshop, Scientific Methods for Understanding Deep Learning arxiv.org/abs/2410.07041

A pure physics paper based on intuitions from AI experiments, expect more of these!

Expériences conduisant aux intuitions. Les résultats des expériences avec des transformers nous ont indiqué où regarder.

I am extremely happy to announce that our paper Can Transformers Do Enumerative Geometry? (arxiv.org/abs/2408.14915) has been accepted to the ICLR 2026 !! Congrats to all my co-authors Alessandro and Roderic Guigo Corominas #ICLR2025 #AI4Math #ORIGINS

All hail the European labor laws!

Open sourcing Int2Int, a Python code base for AI for maths, with a special focus on arithmetic and number theory github.com/f-charton/Int2… A user manual, and instructions on how to extend it, can be found here arxiv.org/abs/2502.17513