Physics AI research that’s shaping the industry. | Mistral AI

What this post added

This post highlights recent research breakthroughs in Physics AI, including papers on "Going with the Speed of Sound: Pushing Neural Surrogates into Highly-turbulent Transonic Regimes" (focusing on 3D wings in transonic regimes with a new dataset), "Fluid Intelligence: A Forward Look on AI Foundation Models in Computational Fluid Dynamics" (deconstructing industrial-scale CFD simulations), "AB-UPT for Automotive and Aerospace Applications" (adding new datasets to AB-UPT use-cases), "GyroSwin: 5D Surrogates for Gyrokinetic Plasma Turbulence Simulations" (addressing plasma turbulence for fusion energy), "AB-UPT" (handling raw geometry without remeshing for aerodynamics CFD), "NeuralDEM" (an end-to-end deep learning surrogate for large-scale multi-physics processes), and "UPT: Universal Physics Transformer" (a framework for scaling neural operators).