Search papers, labs, and topics across Lattice.
Los Alamos National Laboratory
3
0
6
Jointly training on simulation and experimental data unlocks balanced prediction in physical systems, outperforming sequential transfer and preserving crucial simulation-only knowledge.
PDE foundation models trained on fluid dynamics struggle to generalize to material dynamics under extreme loading, highlighting a significant gap in their out-of-distribution capabilities.
A single Latent Evolution Model can solve the trifecta of forward modeling, inverse problems, and tuning for accelerator beam diagnostics, offering a unified approach to traditionally separate challenges.
