你能长多远？表征材料科学中图生成模型的外推前沿
How Far Can You Grow? Characterizing the Extrapolation Frontier of Graph Generative Models for Materials Science
AI总结：
提出RADII基准，通过半径分辨的纳米粒子结构评估晶体生成模型的外推能力，发现模型在训练半径外误差增加，且外推前沿可预测。
摘要：
每种晶体材料生成模型都存在一个临界结构尺寸，超出该尺寸其输出变得不可靠；我们称之为外推前沿。尽管这对纳米材料设计有重要影响，但这一前沿从未被系统测量过。我们引入RADII，一个半径分辨的基准，包含约75,000个晶体衍生的纳米粒子结构（33-11,298个原子），将半径视为连续缩放旋钮，在无泄漏分割下追踪从分布内到分布外的生成质量。每个模型以目标组成和原子数为条件，将几何外推作为评估变量。RADII提供前沿特定的诊断：每个半径的误差曲线精确定位每个架构的缩放上限，表面-内部分解分离边界和体相失效，跨度量排序揭示结构保真度的哪个方面首先失效。对五种最先进架构进行基准测试，我们发现：(i) 表现良好的模型在训练半径外全局位置误差增加约13%，而发散模型在所有尺度上保真度差，局部键合保真度从可忽略的退化到超过2倍的误差增长；(ii) 没有两个架构共享相同的失效序列，揭示前沿是由模型族决定的多维表面；(iii) 表现良好的模型遵循预期的几何缩放指数α ~ 1/3，其分布内拟合可预测分布外误差，使前沿可预测。将MatterGen扩展到其公布的参数数量稳定了采样，但并未关闭前沿，而DiffCSP在公布规模下仍不稳定。这些发现将输出尺度确立为几何生成模型的一级评估轴。代码和数据：https://github.com/KurbanIntelligenceLab/RADII。
英文摘要：
Every generative model for crystalline materials harbors a critical structure size beyond which its outputs become unreliable; we call this the extrapolation frontier. Despite its consequences for nanomaterial design, this frontier has never been systematically measured. We introduce RADII, a radius-resolved benchmark of ~75,000 crystal-derived nanoparticle structures (33-11,298 atoms) that treats radius as a continuous scaling knob, tracing generation quality from in- to out-of-distribution under leakage-free splits. Each model is conditioned on target composition and atom count, isolating geometric extrapolation as the evaluation variable. RADII provides frontier-specific diagnostics: per-radius error profiles pinpoint each architecture's scaling ceiling, surface-interior decomposition separates boundary from bulk failures, and cross-metric sequencing reveals which aspect of structural fidelity breaks first. Benchmarking five state-of-the-art architectures, we find that: (i) well-behaved models degrade by ~13% in global positional error beyond training radii, while divergent models show poor fidelity across scales, with local bond fidelity ranging from negligible degradation to over 2x error growth; (ii) no two architectures share a failure sequence, revealing the frontier as a multi-dimensional surface shaped by model family; and (iii) well-behaved models follow the expected geometric scaling exponent alpha ~ 1/3, whose in-distribution fit predicts out-of-distribution error, making frontiers forecastable. Scaling MatterGen to its published parameter count stabilizes sampling but does not close the frontier, while DiffCSP remains unstable at published scale. These findings establish output scale as a first-class evaluation axis for geometric generative models. Code and data: https://github.com/KurbanIntelligenceLab/RADII.
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。
