ATLAS: A Foundation Neural Sampler for Amorphous Materials
作者: Mouyang Cheng, Denis Blessing, Botao Yu, Gerhard Neumann, Mingda Li, Carles Domingo-Enrich, Yuanqi Du
分类: cond-mat.mtrl-sci, cs.LG, physics.comp-ph
发布日期: 2026-07-21
💡 一句话要点
提出ATLAS以高效采样无定形材料
🎯 匹配领域: 支柱九:具身大模型 (Embodied Foundation Models)
关键词: 无定形材料 图神经网络 采样方法 热力学量 材料设计 高熵金属玻璃 机器学习 扩散过程
📋 核心要点
- 现有的分子动力学和蒙特卡洛方法在低温下效率低下,难以有效采样无定形材料的能量景观。
- ATLAS通过学习扩散过程,利用图神经网络生成无定形结构,并能高效估计热力学量。
- 在实验中,ATLAS在低温玻璃状态下实现了低于0.2%的自由能误差,并有效恢复了短程有序趋势。
📝 摘要(中文)
无定形材料具有优异的机械和功能特性,但其复杂的能量景观使得采样变得困难。在玻璃转变温度以下,传统的分子动力学和蒙特卡洛方法效率低下,而数据驱动的生成模型又受限于稀缺和偏倚的参考集合。本文提出ATLAS,一个高效的采样器,通过学习扩散过程直接从目标能量函数生成玻尔兹曼分布的无定形结构。ATLAS通过一个等变图神经网络进行参数化,能够在系统大小、温度和组成上进行泛化。实验结果表明,ATLAS在低温玻璃状态下的自由能误差低于0.2%,且能量评估次数减少超过500倍。
🔬 方法详解
问题定义:本文旨在解决无定形材料的采样问题,现有方法在低温下因稀有的障碍跨越事件而导致效率低下。
核心思路:ATLAS通过学习扩散过程,直接从目标能量函数生成无定形结构,克服了传统方法的局限性。
技术框架:ATLAS的整体架构包括一个等变图神经网络,能够处理不同大小、温度和组成的系统,并利用扩散过程的时间反转来估计热力学量。
关键创新:ATLAS的核心创新在于其高效的采样机制和对目标可观测量的引导能力,显著减少了能量评估次数。
关键设计:ATLAS采用了composition-amortized预训练策略,优于从头开始的特定组成训练,且在网络结构上使用了图神经网络以处理复杂的材料结构。
🖼️ 关键图片
📊 实验亮点
ATLAS在二维Kob-Andersen系统中成功重现了平行温度马尔可夫链蒙特卡洛的结构分布,自由能和熵的误差低于0.2%,且能量评估次数减少超过500倍。在Cu-Zr和Cr-Co-Ni金属玻璃中,ATLAS有效恢复了实验观察到的短程有序趋势。
🎯 应用场景
ATLAS的研究成果在材料科学领域具有广泛的应用潜力,尤其是在无定形材料的设计与优化方面。其高效的采样能力和对热力学量的准确估计,能够推动新材料的开发,特别是在高熵金属玻璃等领域,具有重要的实际价值和未来影响。
📄 摘要(原文)
Amorphous materials exhibit exceptional mechanical and functional properties, yet their rugged energy landscapes are notoriously difficult to sample. Below the glass-transition temperature, conventional molecular dynamics and Monte Carlo become inefficient because equilibration relies on rare barrier-crossing events, while data-driven generative models are constrained by scarce and biased reference ensembles. Here, we introduce ATLAS, an efficient sampler that learns a diffusion process to generate Boltzmann-distributed amorphous structures directly from a target energy function. Parameterized by an equivariant graph neural network, ATLAS generalizes across system size, temperature, and composition. By exploiting the time reversal of the diffusion process, it enables efficient estimation of thermodynamic quantities and steering toward target observables. In two-dimensional Kob-Andersen systems, ATLAS reproduces parallel tempering Markov chain Monte Carlo structural distributions, free energies and entropies, achieving below 0.2% free energy error in the low-temperature glass regime with over 500-fold fewer energy evaluations. In Cu-Zr and Cr-Co-Ni metallic glasses, ATLAS recovers experimentally observed short-range-order trends and steers structures toward prescribed order parameters and optimized bulk moduli. Moreover, composition-amortized pretraining outperforms composition-specific training from scratch, reduces inverse-design costs by several hundred-fold, and enables sampling with expensive universal machine learning interatomic potentials. Coupled to a large language model agent, ATLAS searches an eight-element space for high-entropy metallic glasses balancing stiffness and ductility, identifying a converged Pareto frontier within 480 oracle evaluations. Together, these results establish ATLAS as a foundation model for sampling, steering and designing amorphous materials.