Enforcing semilinearity via local stencil-scale normalization and training on polynomial profiles produces stable, generalizable neural advection schemes with a new flux limiter that improves shape preservation over OSTVD3.
Learning second- order tvd flux limiters using differentiable solvers.arXiv preprint arXiv:2503.09625, 2025
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
years
2026 2verdicts
UNVERDICTED 2representative citing papers
In a 1D plasma benchmark, classical conservative finite volume structure achieves rollout MSE of 7.35e-9 and wins 60/64 cases, outperforming neural baselines focused on one-step accuracy.
citing papers explorer
-
Design principles for stable and generalizable data-driven discretizations for solving linear hyperbolic conservation laws
Enforcing semilinearity via local stencil-scale normalization and training on polynomial profiles produces stable, generalizable neural advection schemes with a new flux limiter that improves shape preservation over OSTVD3.
-
Conservative Discrete Structure Stabilizes Autoregressive Rollouts in a 1D Drift Diffusion Poisson Benchmark
In a 1D plasma benchmark, classical conservative finite volume structure achieves rollout MSE of 7.35e-9 and wins 60/64 cases, outperforming neural baselines focused on one-step accuracy.