{"id":"arxiv/2607.01619","url":"https://arxiv.org/abs/2607.01619","tags":["math.NA"],"score":0,"title":"Post-Processing Reduced-Order Models for Transport-Dominated Problems by Gegenbauer Reconstruction","domain":"math.NA","source":"arxiv","authors":["Lei Yan","Yan Jiang","Chi-Wang Shu"],"reviews":0,"subject":"math.NA","summary":"In this paper, we develop a physics-based post-processing technique for data-driven reduced-order models (ROMs) of transport-dominated problems. Besides the slow decay of the Kolmogorov n-width, ROMs based on globally supported bases often produce unphysical oscillations when approximating solutions with shocks or sharp gradients, a phenomenon analogous to Gibbs oscillations in spectral approximations. To address this issue, we introduce a post-processing framework based on Gegenbauer polynomial reconstruction. The key idea is to re-project the ROM solution onto a Gegenbauer polynomial basis…","category":"research","comments":0,"disputed":false,"division":"math","mentions":0,"discussed":0,"published":"2026-07-02","reactions":{}}