通过自动诊断与技能发现训练数值智能
原标题：Training Numerical Intelligence via Auto-Diagnosis and Skill Discovery
AI 摘要
该论文提出 ADSD（Auto-Diagnosis and Skill Discovery）框架，将数值求解器的改进从试错式编辑转变为「诊断—发现—实现」的结构化流程：先解释求解器表现不佳的原因，再据此发现合适的数值方法，并封装为可复用的求解器技能。作者在潮流方程、AC 最优潮流控制、刚性常微分方程和异质扩散偏微分方程四个数值领域，用 GPT-5.5、Claude-sonnet-5、Qwen3.8-2.4T 三个智能体基线进行评测，结果显示求解器在精度、鲁棒性和效率上均有提升，例如在 GOC-500 潮流问题上将 Qwen3.8-2.4T 的平均误差降低近一个数量级，并可迁移到未见过的电网拓扑和运行工况。
正文节选
Training Numerical Intelligence via Auto-Diagnosis and Skill Discovery Abstract AI agents are becoming increasingly capable of generating scientific code, but generating code is not the same as improving the algorithms behind it. For numerical solvers, execution feedback can expose poor performance, but rarely reveals its underlying cause and how to address it. We introduce Auto-Diagnosis and Skill Discovery (ADSD), a framework that links numerical diagnosis to reusable solver self-improvement.