cs.AINov 7, 2026

Physics2Weight: Direct Neural Network Architecture and Weight Generation from Physical Laws Using Large Language Models

Authors: Shogo Sato, Tomoyasu Shimada, Kazuhiko Murasaki, Takashi Sano

Organizations: NTT, Inc., Japan.

Abstract

Recent advances in large language models (LLMs) have enabled weight-space programming (WSP), which directly generates neural network weights without gradient-based optimization. However, extending WSP to physical problems remains challenging, particularly in achieving both solution accuracy and physical consistency. In this paper, we propose Physics2Weight (PtW), a framework that extends WSP to solving partial differential equations by directly generating neural network weights and editing architectures based on governing equations and initial and boundary conditions. PtW introduces physics-constrained prompting and physics-guided architecture editing, combined with iterative weight refinement from WSP, to improve solution accuracy and physical consistency without gradient-based optimization. Experiments on four PDEs, namely the heat, advection, Burgers', and Allen--Cahn equations, demonstrate lower solution errors than conventional physics-informed neural networks, while revealing limitations in satisfying the governing equations for nonlinear PDEs.

Figures & tables

Appendix figures & tables2 assets

Supplementary material from the paper’s appendix.

Appendix

Explore similar work

CardsList
  1. Loss-Conditional PINNs for Parametric PDE Families

    Jun 3, 2026Anna Lazareva, Alexander TarakanovParameterized PDE SolvingPDE Solving

  2. An Optimisation Framework for the Well-Conditioned Training of Physics-Informed Neural Networks

    Jul 2, 2026Joseph Webb, Sadok Jerad, Coralia CartisGauss-Newton OptimizationPDE Solving

  3. Physics-Informed Broad Learning System: An Efficient Backpropagation-Free Framework for Solving Partial Differential Equations

    Jul 28, 2026Pinki Khatun, M. Sajid, Abhinav Jha +1Neural PDE SolversPDE Solving