发表机构
Beihang University（北京航空航天大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
CrystalJev将原子级基础模型用作快速决策器，通过单次前向传播以校准概率预测材料稳定性，成本仅为弛豫的三十分之一，并在前瞻测试中误差不超过2.1个百分点。
AI 中文摘要
原子级基础模型对数百万种假设材料进行分类，但通常被用作缓慢的模拟器，其阈值能量被直接采信。它们更适合被解读为快速的决策者。CrystalJev对每个未弛豫结构仅查询一次冻结的原子间势，并以校准概率、有限样本保证以及何时进行慢速思考的规则来回答类型化问题。在65个Matbench Discovery模型中，“稳定”调用实际上是概率的伪装，可通过模型的误差和候选总体来解释。一旦训练完成，一次前向传播的决策效果几乎与弛豫相当，而成本仅为后者的三十分之一；价值信息理论仅在决策可能改变时引导更慢的计算。同一层还可回答电子、力学和分子问题。在一项注册的前瞻性测试中，包含700个新的密度泛函计算，仅基于现有数据校准的单次预测对未见候选者稳定分数（5.8%）的高估最多不超过2.1个百分点。
英文摘要
Atomistic foundation models triage millions of hypothetical materials but are used as slow simulators, their thresholded energies taken at face value. They are better read as fast decision-makers. CrystalJev queries a frozen interatomic potential once per unrelaxed structure and answers typed questions with calibrated probabilities, finite-sample guarantees and a rule for when to think slowly. Across 65 Matbench Discovery models, a 'stable' call is a probability in disguise, explained by a model's errors and the candidate population. Once trained, one forward pass decides nearly as well as a relaxation at a thirtieth of its cost, and a value-of-information theory sends slower computation only where decisions can change. The same layer answers electronic, mechanical and molecular questions. In a registered prospective test with 700 new density-functional calculations, single-pass forecasts calibrated only on existing data over-stated the stable fraction of unseen candidates (5.8%) by at most 2.1 percentage points.
Comments43 pages, 6 main figures, 5 Extended Data figures, 1 Extended Data table; Supplementary Information included
