3 papers
cs.LG2026
MATRIX: A Multimodal Benchmark and Post-Training Framework for Materials Science
Delia McGrath, Curtis Chong, Rohil Kulkarni +2
Scientific reasoning in materials science requires integrating multimodal experimental evidence with underlying physical theory. Existing benchmarks make it difficult to assess whe…
physics.chem-ph2025
Beyond Force Metrics: Pre-Training MLFFs for Stable MD Simulations
Shagun Maheshwari, Zhengxian Tang, Janghoon Ock +3
Machine-learning force fields (MLFFs) have emerged as a promising solution for speeding up ab initio molecular dynamics (MD) simulations, where accurate force predictions are criti…
physics.comp-ph2025
TorchSim: An efficient atomistic simulation engine in PyTorch
Orion Cohen, Janosh Riebesell, Rhys Goodall +6
We introduce TorchSim, an open-source atomistic simulation engine tailored for the Machine Learned Interatomic Potential (MLIP) era. By rewriting core atomistic simulation primitiv…