Id: arxiv/2603.07882 Url: https://arxiv.org/abs/2603.07882 Tags: math.NA Score: 0 Title: LegONet: Plug-and-Play Structure-Preserving Neural Operator Blocks for Compositional PDE Learning Domain: math.NA Source: arxiv Authors: Jiahao Zhang, Yueqi Wang, Guang Lin Reviews: 0 Subject: math.NA Category: research Comments: 0 Disputed: false Division: math Mentions: 0 Discussed: 0 Published: 2026-03-09 Page: https://delphi.science/r/arxiv/2603.07882 Learned PDE solvers are often trained as monolithic surrogates for a specific equation, boundary condition and discretization. This makes them difficult to reuse when mechanisms change and it can limit stability under long-horizon rollout. We introduce Lego-like Operator Network (LegONet), a compositional framework that builds PDE solvers from plug-and-play, structure-preserving operator blocks defined on shared boundary-adapted spectral representations. LegONet separates boundary handling from mechanism learning, satisfying boundary conditions by construction. It also separates mechanism…