Add initial dataset card for MLIP Arena
#1
by
nielsr
HF Staff
- opened
README.md
ADDED
|
@@ -0,0 +1,19 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
task_categories:
|
| 3 |
+
- other
|
| 4 |
+
tags:
|
| 5 |
+
- materials-science
|
| 6 |
+
- molecular-modeling
|
| 7 |
+
- interatomic-potentials
|
| 8 |
+
- benchmark
|
| 9 |
+
---
|
| 10 |
+
|
| 11 |
+
# MLIP Arena Dataset
|
| 12 |
+
|
| 13 |
+
This repository provides access to the data for **MLIP Arena**, a benchmark platform for Machine Learning Interatomic Potentials (MLIPs). It was presented in the paper [MLIP Arena: Advancing Fairness and Transparency in Machine Learning Interatomic Potentials via an Open, Accessible Benchmark Platform](https://huggingface.co/papers/2509.20630).
|
| 14 |
+
|
| 15 |
+
MLIP Arena evaluates force field performance based on physics awareness, chemical reactivity, stability under extreme conditions, and predictive capabilities for thermodynamic properties and physical phenomena. By moving beyond static DFT references, it reveals important failure modes of current foundation MLIPs in real-world settings, providing a reproducible framework to guide next-generation MLIP development toward improved predictive accuracy and runtime efficiency while maintaining physical consistency.
|
| 16 |
+
|
| 17 |
+
- **Paper:** [MLIP Arena: Advancing Fairness and Transparency in Machine Learning Interatomic Potentials via an Open, Accessible Benchmark Platform](https://huggingface.co/papers/2509.20630)
|
| 18 |
+
- **Code:** [https://github.com/atomind-ai/mlip-arena](https://github.com/atomind-ai/mlip-arena)
|
| 19 |
+
- **Project Page:** [https://huggingface.co/spaces/atomind/mlip-arena](https://huggingface.co/spaces/atomind/mlip-arena)
|