Neural operator models trained on simulation data often lose accuracy when applied to experimental measurements due to the sim-to-real gap. Standard fine-tuning with limited real data can reduce this gap, but it may also damage the core physics-relevant representations learned during pretraining. Although knowledge-preserving adaptation has been widely investigated in vision or language tasks, it remains unclear whether these methods are suitable for neural operators whose architectures and protected knowledge are fundamentally different. Neural operators need to preserve core-scale physical structures rather than semantic or visual features. We propose PhysGuard, a physics-preserving framework for accurate sim-to-real adaptation of neural operators. Specifically, PhysGuard uses the empirical Fisher Information Matrix computed on simulation data to identify physics-critical parameter directions, then restricts fine-tuning updates to directions that do not interfere with them. A layer-wise Gram-matrix formulation makes this efficient for models with millions of parameters, while an adaptive threshold automatically determines the protected subspace size. A spectral probe experiment shows that the dominant Fisher directions are strongly associated with low-frequency output structures. Experiments on benchmark across four neural operator architectures and different physical systems show that PhysGuard performs strongly on most evaluation metrics compared to baselines. The benefits are most evident under severe domain shift, where it reduces low-frequency error by up to 32\% compared to standard fine-tuning while maintaining adaptability. Our code is available at https://github.com/ZhouChaunge/PhysGuard.

翻译：基于仿真数据训练的神经算子模型在应用于实验测量时，由于仿真到现实（sim-to-real）的差距常常会损失准确性。使用有限的真实数据进行标准微调可以缩小这一差距，但也可能破坏预训练阶段学习到的核心物理相关表征。尽管知识保留适应方法在视觉或语言任务中已被广泛研究，但对于架构和保护知识存在根本差异的神经算子，这些方法是否适用仍不明确。神经算子需要保留核心尺度的物理结构，而非语义或视觉特征。我们提出PhysGuard，一种面向神经算子精准仿真到现实适应的物理保留框架。具体而言，PhysGuard利用基于仿真数据计算的经验Fisher信息矩阵（Fisher Information Matrix）识别物理关键参数方向，随后将微调更新限制在不干扰这些方向的方向上。通过逐层Gram矩阵公式化，该方法可高效处理百万级参数模型，同时自适应阈值自动确定受保护子空间的大小。频谱探针实验表明，主导Fisher方向与低频输出结构密切相关。在涵盖四种神经算子架构及不同物理系统的基准实验中，PhysGuard在多数评估指标上均优于基线方法。其在严重域偏移场景下的优势最为显著，相较于标准微调，低频误差最高降低32%，同时保持适应性。我们的代码开源在https://github.com/ZhouChaunge/PhysGuard。