Add model card

README.md

@@ -1,3 +1,151 @@
 ---
 license: apache-2.0
+tags:
+  - physics-ai
+  - pde
+  - neural-pde-solvers
+  - surrogate-modeling
+  - 3d-geometry
+  - scientific-machine-learning
+  - neurips-2026
+arxiv: 2610.03363
+pipeline_tag: feature-extraction
 ---
+
+<h1 align="center">
+  <br>Geometry Meets Physics (GMP)<br>
+</h1>
+<h4 align="center">Data-Efficient Pre-Training for Unstructured Neural PDE Solvers</h4>
+<p align="center"><i>Accepted at NeurIPS 2026</i></p>
+
+<p align="center">
+<a href="https://arxiv.org/abs/2610.03363">
+  <img src="https://img.shields.io/badge/arXiv-2610.03363-b31b1b?logo=arxiv" alt="Read on arXiv"/>
+</a>
+<a href="https://github.com/tum-pbs/GMP">
+  <img src="https://img.shields.io/badge/GitHub-tum--pbs%2FGMP-181717?logo=github" alt="GitHub Repository"/>
+</a>
+<a href="https://huggingface.co/thuerey-group/GMP">
+  <img src="https://img.shields.io/badge/Huggingface-GMP-FFD21E?logo=huggingface" alt="Pretrained weights on Hugging Face"/>
+</a>
+<a href="LICENSE">
+  <img src="https://img.shields.io/badge/License-Apache%202.0-blue.svg" alt="License: Apache 2.0"/>
+</a>
+</p>
+
+<p align="center">
+  <img src="figures/drivaerml_error_comparison.png" width="49%" alt="DrivAerML Error Comparison" />
+  <img src="figures/shiftwing_error_comparison.png" width="49%" alt="SHIFT-Wing Error Comparison" />
+</p>
+
+---
+
+## 📌 Overview
+
+**GMP (Geometry Meets Physics)** is a disk-data-free pre-training framework for neural surrogate models on unstructured 3D geometries, tailored to both steady-state and transient physical regimes.
+
+Instead of relying on massive, pre-computed simulation data stored on disk (which often consume terabytes of storage and suffer from severe I/O bottlenecks), GMP generates pre-training data **purely on-the-fly**:
+- **Steady-state regime (Geometry-driven):** Employs intrinsic shape descriptors (coordinates, curvatures, normals, and SGD) to learn rich representations of complex 3D domains without running expensive offline simulations.
+- **Transient regime (Physics-driven):** Leverages GPU-accelerated solvers to generate synthetic PDE dynamics online, enabling scalable pre-training for time-dependent evolutions without storage overhead.
+
+For full code, benchmark datasets, and evaluation scripts, visit the official **[GitHub Repository](https://github.com/tum-pbs/GMP)**.
+
+---
+
+## 📦 Pre-Trained Checkpoints
+
+This repository hosts official pre-trained checkpoints for **GMP-Net** and baseline architectures:
+
+| Model / Baseline | Pre-Training Stage | Regime | Checkpoint Path in Repo | Size |
+| :--- | :--- | :--- | :--- | :--- |
+| **GMP-Net** | Geometry Pre-Training | Steady-state | `gmpnet/epoch-200.pth` | ~346 MB |
+| **GMP-Net** | Physics Pre-Training | Transient | `gmpnet_physics/epoch-199-EMA.ckpt` | ~174 MB |
+| **AB-UPT** | Geometry Pre-Training | Steady-state | `abupt/epoch-200.pth` | ~148 MB |
+| **Transolver++** | Geometry Pre-Training (Surface) | Steady-state | `transolver_pp_surface/epoch-200.pth` | ~106 MB |
+| **Transolver++** | Geometry Pre-Training (Volume) | Steady-state | `transolver_pp_volume/epoch-200.pth` | ~106 MB |
+
+---
+
+## 🚀 How to Download the Weights
+
+### Option 1: Download All Checkpoints into `outputs_pretraining/` (Recommended)
+
+Using Python:
+```python
+from huggingface_hub import snapshot_download
+
+# Downloads all model weights directly into the folder expected by the GMP codebase
+snapshot_download(
+    repo_id="thuerey-group/GMP",
+    local_dir="outputs_pretraining",
+    allow_patterns=["*.pth", "*.ckpt"],
+)
+```
+
+Or with the `hf` CLI:
+```bash
+hf download thuerey-group/GMP --local-dir outputs_pretraining --include "*.pth" --include "*.ckpt"
+```
+
+Once downloaded, the directory matches the expected hierarchy in the codebase:
+```text
+outputs_pretraining/
+├── gmpnet/
+│   └── epoch-200.pth
+├── gmpnet_physics/
+│   └── epoch-199-EMA.ckpt
+├── abupt/
+│   └── epoch-200.pth
+├── transolver_pp_surface/
+│   └── epoch-200.pth
+└── transolver_pp_volume/
+    └── epoch-200.pth
+```
+
+### Option 2: Download a Specific Checkpoint
+
+```python
+from huggingface_hub import hf_hub_download
+
+checkpoint_path = hf_hub_download(
+    repo_id="thuerey-group/GMP",
+    filename="gmpnet/epoch-200.pth",
+    local_dir="outputs_pretraining",
+)
+```
+
+---
+
+## 🔬 Supported Downstream Benchmarks
+
+The pre-trained models can be directly fine-tuned across six downstream engineering benchmarks on unstructured surface and volumetric meshes:
+- **DrivAerML**: Automotive aerodynamics (3D Surface & Volume).
+- **SHIFT-Wing**: Aerospace aerodynamics (NASA CRM wings under varying Mach/AoA).
+- **SHIFT-Crash (Steady)**: Vehicle structural deformation and Von Mises stress.
+- **Kuramoto–Sivashinsky (KS)**: 2D chaotic spatiotemporal dynamics.
+- **Flow Around Ellipse**: 2D unsteady fluid mechanics past an elliptical obstacle.
+- **SHIFT-Crash (Transient)**: Autoregressive collision and impact dynamics rollout across 12 timesteps.
+
+Refer to the [GitHub Repository](https://github.com/tum-pbs/GMP) for Hydra fine-tuning commands.
+
+---
+
+## 📜 Citation
+
+If you use these pre-trained weights or the GMP framework in your research, please cite our paper:
+
+```bibtex
+@inproceedings{medrano2026geometry,
+  title={Geometry Meets Physics: Data-Efficient Pre-Training for Unstructured Neural PDE Solvers},
+  author={Medrano-Navarro, Luis and Baldan, Giacomo and Liu, Qiang and Holzschuh, Benjamin and Hagnberger, Jan and Niepert, Mathias and Thuerey, Nils},
+  booktitle={The Fortieth Annual Conference on Neural Information Processing Systems},
+  year={2026},
+  url={https://openreview.net/forum?id=jidnQA2TIz}
+}
+```
+
+---
+
+## 📄 License
+
+The checkpoints and code are released under the [Apache 2.0 License](LICENSE).