4 papers
Exascale Multi-Task Graph Foundation Models for Imbalanced, Multi-Fidelity Atomistic Data
Massimiliano Lupo Pasini, Jong Youl Choi, Kshitij Mehta +8
We present an exascale workflow for materials discovery using atomistic graph foundation models built on HydraGNN. We jointly train on 16 open first-principles datasets (544+ milli…
Multi-task parallelism for robust pre-training of graph foundation models on multi-source, multi-fidelity atomistic modeling data
Massimiliano Lupo Pasini, Jong Youl Choi, Pei Zhang +6
Graph foundation models using graph neural networks promise sustainable, efficient atomistic modeling. To tackle challenges of processing multi-source, multi-fidelity data during p…
Real-time interpretation of neutron vibrational spectra with symmetry-equivariant Hessian matrix prediction
Bowen Han, Pei Zhang, Kshitij Mehta +3
The vibrational behavior of molecules serves as a crucial fingerprint of their structure, chemical state, and surrounding environment. Neutron vibrational spectroscopy provides com…
Scalable Training of Trustworthy and Energy-Efficient Predictive Graph Foundation Models for Atomistic Materials Modeling: A Case Study with HydraGNN
Massimiliano Lupo Pasini, Jong Youl Choi, Kshitij Mehta +8
We present our work on developing and training scalable, trustworthy, and energy-efficient predictive graph foundation models (GFMs) using HydraGNN, a multi-headed graph convolutio…