{"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","summary":"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…","category":"research","comments":0,"disputed":false,"division":"math","mentions":0,"discussed":0,"published":"2026-03-09","reactions":{}}