从基础机器学习原子间势中提取潜静电
Distilling latent electrostatics from foundation machine learning interatomic potentials
- Department of Chemistry, UC Berkeley（加州大学伯克利分校化学系）
- Bakar Institute of Digital Materials for the Planet, UC Berkeley（伯克利大学数字材料研究所）
- Chemical Sciences Division, Lawrence Berkeley National Laboratory（伯克利国家实验室化学科学部）
机构由 AI 辅助整理，请以论文原文为准。
AI总结：
提出潜埃瓦德求和（LES）方法，从基础机器学习原子间势中提取潜静电，训练轻量级学生模型，降低计算成本并提供玻恩有效电荷张量和红外光谱，基准测试表明教师模型的DFT级别比架构更重要。
AI中文摘要：
基础机器学习原子间势（MLIPs）已能够在化学和材料空间的广泛区域进行原子模拟，但许多模型计算成本高昂且缺乏显式静电，限制了其在长程相互作用和电响应主导的系统中的应用。此前，我们引入了潜埃瓦德求和（LES），该方法仅从密度泛函理论（DFT）能量和力标签中学习潜原子电荷和长程静电。在此，我们使用LES提取基础模型中潜藏的静电：教师模型预测的能量和力用于训练轻量级LES增强的学生MLIP，并可选择在额外DFT数据上进行微调。所得模型降低了计算成本，同时提供了玻恩有效电荷张量和红外光谱。我们针对液态水、浓盐酸和锐钛矿TiO2(101)-水界面的实验红外光谱，对从多种基础MLIP（包括基于UMA、MACE、Orb、eSEN、GemNet-OC、PET和EquiformerV2的模型）蒸馏得到的学生模型进行了基准测试。在这些系统中，大多数基础MLIP都能提取静电响应。基准测试进一步表明，用于训练教师模型的底层DFT水平和数据集在决定静电和光谱精度方面比架构更重要。对于TiO2-水界面，使用适量更高级别DFT数据进行微调可改善结构和红外预测。因此，基于LES的蒸馏提供了一条实用途径，将基础MLIP转化为高效、电响应的模型，同时测试了基础模型中编码的物理保真度。
英文摘要：
Foundation machine learning interatomic potentials (MLIPs) have enabled atomistic simulations across broad regions of chemical and materials space, but many remain computationally expensive and lack explicit electrostatics, limiting their use for systems governed by long-range interactions and electrical response. Previously, we introduced Latent Ewald Summation (LES), which learns latent atomic charges and long-range electrostatics from density functional theory (DFT) energy and force labels alone. Here, we use LES to extract electrostatics that are latent in foundation models: energies and forces predicted by a teacher model are used to train a lightweight LES-augmented student MLIP, with optional fine-tuning on additional DFT data. The resulting models reduce computational cost while providing access to Born effective charge tensors, and infrared spectra. We benchmark student models distilled from a broad set of foundation MLIPs, including UMA, MACE, Orb, eSEN, GemNet-OC, PET, and EquiformerV2-based models, against experimental infrared spectra for liquid water, concentrated hydrochloric acid, and the anatase TiO2(101)-water interface. Across these systems, electrostatic response can be extracted from most foundation MLIPs. The benchmark further shows that the underlying DFT level and dataset used to train the teacher model play a larger role than architecture in determining electrostatic and spectroscopic accuracy. For the TiO2-water interface, fine-tuning with a modest amount of higher-level DFT data improves structural and infrared predictions. LES-based distillation therefore provides a practical route for converting foundation MLIPs into efficient, electrically responsive models, while also testing the physical fidelity encoded in foundation models.
