Search papers, labs, and topics across Lattice.
Tsinghua AIAI for Science InstituteGaoling AISchool of Mechanics and EngineeringTJUMay 25, 2026arXiv:2605.25786The paper introduces NPSolver, a neural operator for solving Poisson equations on irregular domains, trained via iterative physics supervision using a few steps of preconditioned conjugate gradient (PCG) to refine its predictions. This iterative refinement provides a more stable and well-scaled training signal compared to using raw PDE residuals or fully converged solutions. The method also incorporates a Boundary-Aware Transolver architecture to better capture boundary conditions, leading to superior performance compared to physics-informed and data-driven baselines on 2D and 3D geometries.
Solving Poisson equations just got faster and more stable: NPSolver trains neural operators without solution labels by iteratively refining predictions with preconditioned conjugate gradient steps.
Efficiently solving Poisson equations on complex, irregular domains remains a fundamental challenge in scientific computing, as classical iterative solvers often suffer from prohibitive runtime due to ill-conditioned systems. While neural operators offer a fast alternative, they typically rely on large-scale labeled datasets or struggle with unstable training dynamics when using physics-informed residual losses. We propose \textsc{NPSolver}, a neural Poisson solver trained without solution labels via iterative physics supervision. Instead of relying on fully converged numerical solutions or raw PDE residuals, \textsc{NPSolver} utilizes a small number of preconditioned conjugate gradient (PCG) steps to refine its own predictions, providing a more stable and well-scaled training signal. Theoretical analysis confirms that this iterative supervision serves as a well-conditioned error proxy and that a stop-gradient design is essential for optimization stability. To better capture boundary-driven features under mixed boundary conditions, we further introduce the Boundary-Aware Transolver (\textsc{BA-Transolver}) architecture that explicitly separates interior and boundary tokenization. Extensive evaluations on 2D and 3D irregular geometries demonstrate that \textsc{NPSolver} outperforms both physics-informed and data-driven baselines. Furthermore, a downstream thermal control task highlights the model's capability for conducting efficient and reliable gradient-based boundary control. We will release our codes and data at https://github.com/intell-sci-comput/NPSolver.
