Showing cs.LGShow all
2 papers · 1 filter
cs.LG2026
Physics-Integrated Operator Learning via Gaussian Splatting Representations
Jihao Zhang, Junyi Guo, Jian-Xun Wang
Neural operators provide efficient surrogates for spatiotemporal PDE systems, but purely data-driven formulations often accumulate substantial errors during long-horizon autoregres…
cs.LG2026
D-Flow SGLD: Source-Space Posterior Sampling for Scientific Inverse Problems with Flow Matching
Meet Hemant Parikh, Yaqin Chen, Jian-Xun Wang
Data assimilation and scientific inverse problems require reconstructing high-dimensional physical states from sparse and noisy observations, ideally with uncertainty-aware posteri…