发表机构
The Beijing National Laboratory for Condensed Matter Physics, Institute of Physics, Chinese Academy of Sciences; Shanghai Artificial Intelligence Laboratory; Dept. of Comp. Sci. and Tech., Institute for AI, Tsinghua University; Institute for AIR, Tsinghua University; Gaoling School of Artificial Intelligence, Renmin University of China; Engineering Research Center of Next-Generation Intelligent Search and Recommendation, MOE; Beijing Key Laboratory of Big Data Management and Analysis Methods; School of Computer Science, Shanghai Jiao Tong University; Zhongguancun Academy（中国科学院物理研究所凝聚态物理国家实验室; 上海人工智能实验室; 清华大学智能产业研究院计算机科学与技术系; 清华大学智能产业研究院; 中国人民大学高瓴人工智能学院; 教育部下一代智能搜索与推荐工程研究中心; 北京大数据管理与分析方法重点实验室; 上海交通大学计算机科学技术学院; 中关村学院）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本文提出生成式相位求解方法PhiGen，利用不变关系学习相位，在210个空间群中高比例恢复中心对称与非中心对称结构图谱，并成功应用于低分辨率沸石数据，为结构测定提供新途径。
AI 中文摘要
晶体结构测定需要散射波的相位，然而衍射仅测量其强度。直接法利用相位不变关系，但随着衍射信息的减少，其可靠性下降。学习型相位预测降低了分辨率障碍，但仍主要局限于具有二元相位的中心对称晶体。我们提出了PhiGen，一种对传统直接法的生成式重构，它学习与原点无关的相位关系，用于二元和连续相位求解。在210个空间群中，包括训练中未出现的空间群，它为99.0%的中心对称结构恢复了高质量图谱，并为92.8%的非中心对称结构恢复了不变性一致的相位。从模拟的3 Å沸石粉末数据中，该网络为84.2%的保留结构恢复了骨架图谱，而Superflip仅为1.0%。对于实验性的ZSM-25和TNU-9，生成的相位为高分辨率相位扩展提供了种子。这些结果表明，从低分辨率、不完整和重叠的衍射数据中确定结构是可行的。
英文摘要
Crystal structure determination requires the phases of scattered waves -- yet diffraction measures only their intensities. Direct methods exploit phase invariants but become less reliable as diffraction information diminishes. Learned phase prediction has lowered the resolution barrier, yet remains primarily confined to centrosymmetric crystals with binary phases. We introduce PhiGen, a generative reformulation of traditional direct methods that learns origin-independent phase relationships for binary and continuous phasing. Across 210 space groups, including groups absent from training, it recovered high-quality maps for 99.0% of centrosymmetric structures and invariant-consistent phases for 92.8% of non-centrosymmetric structures. From simulated 3 Å zeolite powder data, the network recovered framework maps for 84.2% of held-out structures, versus 1.0% for Superflip. For experimental ZSM-25 and TNU-9, generated phases seeded high-resolution phase extension. These results suggest a route to structure determination from low-resolution, incomplete, and overlapped diffraction data.
