arXiv cs.AI· Sergio Rinc\'on, Gabriel Gonz\'alez, Nicol\'as Andrade, Rafael Vel\'asquez, Sebasti\'an Ojeda, Juanita Puentes, Nicol\'as Aparicio, Paula C\'ardenas, Pablo Arbel\'aez·· 3 小时前AI 评分18
OPAL-Bench：晶体结构预测的定量评估基准
Quantitative Assessment of Crystal Structure Prediction
AI 导读
研究者推出 OPAL-Bench 基准，将晶体结构预测（CSP）与对称性信息 CSP（SICSP）作为独立任务评估，并发布 SE(3) 等变流匹配基线 OPAL-CSP。该基准在 4 个数据集上用无阈值几何与物理距离指标评测了 7 个 CSP 方法和 2 个 SICSP 方法。结果显示没有任何方法在所有指标上占优，几何性能与物理合理性也不强相关。
正文 · 原文
Abstract:Crystal Structure Prediction (CSP) aims to predict a material's 3D structure from its chemical composition, with applications from energy storage to biomedical devices. Despite its importance, existing evaluation frameworks suffer from key limitations: they rely on threshold-sensitive or incomparable metrics, focus predominantly on geometric performance while overlooking physical plausibility, and fail to account for polymorphism, where a chemical composition can crystallize into multiple stable structures. In parallel, recent Symmetry-Informed CSP (SICSP) methods often report results alongside standard CSP baselines while receiving crystallographic templates at inference, which constrain the search space and make SICSP a fundamentally different task. To address these limitations, we introduce an Orientation-invariant, Polymorph-Aware, Large Benchmark (OPAL-Bench) that evaluates CSP and SICSP as separate tasks. We evaluate 7 CSP and 2 SICSP methods under OPAL-Bench across 4 datasets using threshold-free geometric and physical distance metrics. Our results show that no method dominates across all metrics, and that geometric performance is not strongly correlated with physical plausibility. We also release OPAL-CSP, an SE(3)-equivariant flow-matching baseline for the CSP track that performs competitively on geometric metrics and ranks among the best on physical ones. By unifying geometric and physical metrics under a polymorph-aware protocol, OPAL-Bench enables a more comprehensive assessment of generative quality in CSP.
| Comments: | Accepted at NeurIPS 2026, Track on Evaluations and Datasets. 49 pages, 21 figures, 17 tables. Code: this https URL |
| Subjects: | Machine Learning (cs.LG); Materials Science (cond-mat.mtrl-sci); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2611.10380 [cs.LG] |
| (or arXiv:2611.10380v2 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2611.10380 arXiv-issued DOI via DataCite |
Submission history
From: Gabriel González [view email]
[v1]
Sat, 7 Nov 2026 01:34:05 UTC (16,613 KB)
[v2]
Tue, 10 Nov 2026 20:15:46 UTC (16,611 KB)
来源：arXiv cs.AI · arxiv.org