![[2602.17176] Universal Fine-Grained Symmetry Inference and Enforcement for Rigorous Crystal Structure Prediction](https://arxiv.org/static/browse/0.3.4/images/arxiv-logo-fb.png){kind=logo}
[2602.17176] Universal Fine-Grained Symmetry Inference and Enforcement for Rigorous Crystal Structure Prediction
Summary
This paper presents a novel approach to crystal structure prediction by utilizing large language models for fine-grained symmetry inference, enhancing the discovery of new materials.
Why It Matters
The research addresses limitations in existing crystal structure prediction methods that rely on historical data. By leveraging advanced AI techniques, this work opens new pathways for material discovery, which is crucial for advancements in various scientific fields, including materials science and chemistry.
Key Takeaways
- Introduces a method that uses large language models for crystal structure prediction.
- Circumvents limitations of database-reliant approaches by generating Wyckoff patterns directly from composition.
- Achieves state-of-the-art performance in stability, uniqueness, and novelty benchmarks.
- Incorporates domain knowledge through constrained optimization for algebraic consistency.
- Expands the exploration of crystallographic space without reliance on existing structures.
Condensed Matter > Materials Science arXiv:2602.17176 (cond-mat) [Submitted on 19 Feb 2026] Title:Universal Fine-Grained Symmetry Inference and Enforcement for Rigorous Crystal Structure Prediction Authors:Shi Yin, Jinming Mu, Xudong Zhu, Lixin He View a PDF of the paper titled Universal Fine-Grained Symmetry Inference and Enforcement for Rigorous Crystal Structure Prediction, by Shi Yin and 3 other authors View PDF HTML (experimental) Abstract:Crystal structure prediction (CSP), which aims to predict the three-dimensional atomic arrangement of a crystal from its composition, is central to materials discovery and mechanistic understanding. Existing deep learning models often treat crystallographic symmetry only as a soft heuristic or rely on space group and Wyckoff templates retrieved from known structures, which limits both physical fidelity and the ability to discover genuinely new material structures. In contrast to retrieval-based methods, our approach leverages large language models to encode chemical semantics and directly generate fine-grained Wyckoff patterns from composition, effectively circumventing the limitations inherent to database lookups. Crucially, we incorporate domain knowledge into the generative process through an efficient constrained-optimization search that rigorously enforces algebraic consistency between site multiplicities and atomic stoichiometry. By integrating this symmetry-consistent template into a diffusion backbone, our approach constrain...
