Machine-learned interatomic potentials (MLIPs) are increasingly used to replace computationally demanding electronic-structure calculations to model matter at the atomic scale. The most commonly used model architectures are constrained to fulfill a number of physical laws exactly, from geometric symmetries to energy conservation. Evidence is mounting that relaxing some of these constraints can be beneficial to the efficiency and (somewhat surprisingly) accuracy of MLIPs, even though care should be taken to avoid qualitative failures associated with the breaking of physical symmetries. Given the recent trend of \emph{scaling up} models to larger numbers of parameters and training samples, a very important question is how unconstrained MLIPs behave in this limit. Here we investigate this issue, showing that -- when trained on large datasets -- unconstrained models can be superior in accuracy and speed when compared to physically constrained models. We assess these models both in terms of benchmark accuracy and in terms of usability in practical scenarios, focusing on static simulation workflows such as geometry optimization and lattice dynamics. We conclude that accurate unconstrained models can be applied with confidence, especially since simple inference-time modifications can be used to recover observables that are consistent with the relevant physical symmetries.

翻译：机器学习原子间势函数（MLIPs）正日益被用于替代计算需求较高的电子结构计算，以在原子尺度上模拟物质。最常用的模型架构受到严格约束，必须精确满足从几何对称性到能量守恒等一系列物理定律。越来越多的证据表明，放宽其中部分约束可能有利于提升MLIPs的效率和（令人惊讶地）准确性，尽管仍需注意避免因破坏物理对称性而导致的定性失效。鉴于当前模型正朝着增加参数规模和训练样本量的方向扩展，一个至关重要的问题是：无约束MLIPs在此极限下的表现如何？本文针对该问题展开研究，结果表明——当基于大规模数据集训练时——无约束模型在精度和速度上均可优于物理约束模型。我们通过基准测试精度和实际应用场景（重点关注几何优化和晶格动力学等静态模拟工作流程）中的可用性两方面评估这些模型。最终得出结论：准确的无约束模型可以放心应用，特别是因为通过简单的推理时修正即可恢复与相关物理对称性一致的观测结果。