arXiv：cs.AI· Peng Kang, Zhen Li, Yu Liu, Lei Zheng, Huibin Xu·· 6 小时前AI 评分42
CrystalJev：用原子级基础模型「快思考与慢思考」加速材料发现
CrystalJev: thinking fast and slow with atomistic foundation models for materials discovery
AI 导读
CrystalJev 将冻结的原子间势模型当作快速决策者，对未弛豫结构单次前向推理即可给出带校准概率和有限样本保证的分类答案，性能接近完整弛豫而成本仅为三十分之一。
正文
Abstract: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.
| Comments: | 43 pages, 6 main figures, 5 Extended Data figures, 1 Extended Data table; Supplementary Information included |
| Subjects: | Materials Science (cond-mat.mtrl-sci); Artificial Intelligence (cs.AI); Computational Physics (physics.comp-ph) |
| Cite as: | arXiv:2610.06985 [cond-mat.mtrl-sci] |
| (or arXiv:2610.06985v1 [cond-mat.mtrl-sci] for this version) | |
| https://doi.org/10.48550/arXiv.2610.06985 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Peng Kang PhD [view email]
[v1]
Sun, 4 Oct 2026 06:20:21 UTC (697 KB)
来源：arXiv：cs.AI · arxiv.org