Graph neural networks can approximate full 3D non-LTE Ca II populations in solar models with correlations above 0.99 and extreme computational efficiency.
Title resolution pending
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
fields
astro-ph.SR 2years
2026 2verdicts
UNVERDICTED 2representative citing papers
Stokes inversions show the Evershed flow reverses to inflow in the upper photosphere while the moat flow remains an outflow, and umbral flashes exhibit supersonic upflows with Mach numbers at least 1.5 consistent with shock fronts.
citing papers explorer
-
Accelerating 3D Non-LTE Synthesis with Graph Neural Networks
Graph neural networks can approximate full 3D non-LTE Ca II populations in solar models with correlations above 0.99 and extreme computational efficiency.
-
Chromospheric and photospheric properties of sunspots as inferred from Stokes inversions under magneto-hydrostatic and non-local-thermodynamic equilibrium
Stokes inversions show the Evershed flow reverses to inflow in the upper photosphere while the moat flow remains an outflow, and umbral flashes exhibit supersonic upflows with Mach numbers at least 1.5 consistent with shock fronts.