arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2503.02407cond-mat.mtrl-scics.LGphysics.comp-ph
收藏

Wyckoff Transformer:对称晶体的生成

Wyckoff Transformer: Generation of Symmetric Crystals

Institute for Functional Intelligent Materials University of Singapore(功能智能材料研究所新加坡大学);School of Materials Science and Engineering(材料科学与工程学院);Nanyang Technological University(南洋理工大学);Constructor University(Constructor大学);Constructor Knowledge Labs(Constructor知识实验室);HSE University(莫斯科高等经济大学);Institute of Materials Research and Engineering(材料研究与工程研究所)

Nikita Kazeev, Wei Nong, Ignat Romanov, Ruiming Zhu, Andrey Ustyuzhanin, Shuya Yamazaki, Kedar Hippalgaonkar

AI总结:

WyFormer基于空间群对称性和Wyckoff位置表示,用置换不变自回归Transformer生成对称有效的晶体结构,兼具高稳定性、准确性质预测和极快推理速度。

摘要:

晶体对称性在决定其物理、化学和电子性质(如电导率和热导率、光学与极化行为以及机械强度)方面起着根本性作用。几乎所有已知的晶体材料都具有内部对称性。然而,现有的生成模型往往未能充分处理这一特性,使得一致地生成稳定且对称性有效的晶体结构成为一个重大挑战。我们引入了WyFormer,一种生成模型,通过正式以空间群对称性为条件直接解决这一问题。它利用Wyckoff位置作为基础,实现了一种优雅、紧凑且离散的结构表示。为了对分布进行建模,我们开发了一种基于Transformer编码器且不使用位置编码的置换不变自回归模型。大量实验表明,WyFormer具有引人注目的属性组合:它实现了同类最佳的对称性条件生成,融入了物理启发的归纳偏置,产生具有竞争性稳定性的结构,即使没有原子坐标也能以竞争性精度预测材料性质,并展现出无与伦比的推理速度。

英文摘要:

Crystal symmetry plays a fundamental role in determining its physical, chemical, and electronic properties such as electrical and thermal conductivity, optical and polarization behavior, and mechanical strength. Almost all known crystalline materials have internal symmetry. However, this is often inadequately addressed by existing generative models, making the consistent generation of stable and symmetrically valid crystal structures a significant challenge. We introduce WyFormer, a generative model that directly tackles this by formally conditioning on space group symmetry. It achieves this by using Wyckoff positions as the basis for an elegant, compressed, and discrete structure representation. To model the distribution, we develop a permutation-invariant autoregressive model based on the Transformer encoder and an absence of positional encoding. Extensive experimentation demonstrates WyFormer's compelling combination of attributes: it achieves best-in-class symmetry-conditioned generation, incorporates a physics-motivated inductive bias, produces structures with competitive stability, predicts material properties with competitive accuracy even without atomic coordinates, and exhibits unparalleled inference speed.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑