CERN and Mistral AI have entered into a long-term partnership aimed at developing fundamental models for scientific research. The collaboration will initially focus on the field of high-energy physics.
The core initiative is the development of a model called Flare, which is categorized as a Large Physics Model. The training process will utilize billions of simulated proton-proton collisions generated by the Large Hadron Collider.
The project aims to create a unified system capable of reconstructing collision parameters, identifying particles, detecting anomalies, and assisting in the design of future detectors. Currently, this process relies on a chain of specialized algorithms.
CERN plans to combine its expertise in particle physics and large-scale computing with Mistral's experience in training large-scale models. Enrica Porcari, head of the IT sector at CERN, noted that high-energy physics provides a rigorous environment for testing AI.
The organization expects that the resulting technologies will eventually be applicable to a broader scientific community beyond particle physics.
