Transferable features in neural fields enable amortized fitting of spatiotemporal scientific signals, reducing iterations by up to 10x while improving reconstruction quality and physical accuracy metrics like gradients and vorticity.
Model Considerations In addition toSIREN[15] andK-Planes[6], we use a modifiedfhashencoder [18] as our spatial encoding-based model
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Fast Amortized Fitting of Scientific Signals Across Time and Ensembles via Transferable Neural Fields
Transferable features in neural fields enable amortized fitting of spatiotemporal scientific signals, reducing iterations by up to 10x while improving reconstruction quality and physical accuracy metrics like gradients and vorticity.