PIMSM: Physics-Informed Multi-Scale Mamba for Stable Neural Representations under Distribution Shift

In three linesPIMSM embeds physical constraints into a multi-scale Mamba architecture to improve representation stability under distribution shift. The model aligns discretization parameters to characteristic frequencies in temporal series (fMRI, weather). Results: improved robustness on Human Connectome Project and Weather-5K with minimal MAE in out-of-distribution forecasting.

Summary generated by Claude — human-verified