CERN and Mistral AI Plan Foundation Model for Collider Physics

CERN and Mistral AI announced in a press release that they signed a statement of intent for a long term partnership focused on foundation models for scientific discovery. Their initial project will create a Large Physics Model trained on billions of simulated proton collisions from the Large Hadron Collider.

CERN describes it as the first foundation model for collider physics. The model will support tasks including reconstructing particle collisions, identifying particles, detecting unexpected signals and evaluating future detector designs.

The project could replace parts of the current reconstruction process, which uses a chain of specialized algorithms to turn detector signals into data for physics analysis. CERN has stored about one exabyte of experimental data from the collider and expects collision volumes to increase tenfold when its upgraded collider begins operating around 2030.