From c532dd21a8448d3ebd5cc67c096cc4a7b3b74308 Mon Sep 17 00:00:00 2001
From: Benedikt Alkin <39829911+BenediktAlkin@users.noreply.github.com>
Date: Mon, 19 Feb 2024 17:39:05 +0100
Subject: [PATCH] Create README.md

---
 README.md | 30 ++++++++++++++++++++++++++++++
 1 file changed, 30 insertions(+)
 create mode 100644 README.md

diff --git a/README.md b/README.md
new file mode 100644
index 0000000..04db532
--- /dev/null
+++ b/README.md
@@ -0,0 +1,30 @@
+[[Project Page](https://ml-jku.github.io/UPT)]
+
+
+**U**niversal **P**hysics **T**ransformers (UPTs) are a novel learning paradigm that can model a wide range of 
+spatio-temporal problems - both for Lagrangian and Eulerian discretization schemes.
+
+
+<p align="center">
+<img width="100%" alt="mimrefiner_schematic" src="https://raw.githubusercontent.com/ml-jku/UPT/main/.github/imgs/schematic.png">
+</p>
+
+
+UPTs can model transient flow simulations (Eulerian discretization scheme) as indicated by test loss and rollout performance (measured via correlation time):
+
+<p align="center">
+<img width="50%" alt="maect_schematic" src="https://raw.githubusercontent.com/ml-jku/UPT/main/.github/imgs/cfd_scaling_testloss.svg">
+<img width="50%" alt="lowshot_vitl" src="https://raw.githubusercontent.com/ml-jku/UPT/main/.github/imgs/cfd_scaling_corrtime.svg">
+</p>
+
+
+<p align="center">
+<img width="100%" alt="mimrefiner_schematic" src="https://raw.githubusercontent.com/ml-jku/UPT/main/.github/imgs/cfd_rollout.png">
+</p>
+
+
+and also the flow-field of particle based simulations (Lagrangian discretization scheme):
+
+<p align="center">
+<img width="100%" alt="mimrefiner_schematic" src="https://raw.githubusercontent.com/ml-jku/UPT/main/.github/imgs/lagrangian_field.png">
+</p>
\ No newline at end of file