**摘要**
Neural surrogate models offer fast approximate mappings from PDE parameters to solutions, but they typically treat solving as a purely statistical task: once trained, they struggle to correct their own constraint violations and extrapolate beyond the training distribution. Recent hybrid methods promote physical correctness by targeting the PDE residual via gradient descent or Gauss--Newton steps,
👤 作者: Haina Jiang, Liam Wang, Peng-Chen Chen, Min Seop Kwak, Seungryong Kim, Brian Bell, Jeong Joon Park

---
🔗 **[Error-Conditioned Neural Solvers](https://arxiv.org/abs/2606.27354v1)**

> Error-Conditioned Neural Solvers
🏷️ 来源: ArXiv cs.AI
⏱️ 2026-06-26 14:00