摘要
高通量X射线衍射(X-ray diffraction,XRD)分析在加速材料发现方面至关重要,但传统方法通常依赖大量人工解释,且在处理复杂的XRD数据时容易忽视低强度峰值信息,从而限制准确性的提升.为解决这一问题,本文提出了一个面向多模态晶体结构预测的大语言模型(large language model,LLM)代理框架,该框架集成了GPT-4驱动的智能代理以及基于XRD和对分布函数的多模态投票模型,能够自主执行晶体结构和空间群预测任务.此外,本文通过引入知识图谱来增强LLM的推理能力,帮助其理解晶体特征之间的关系,进一步提升预测的准确性.实验结果表明,该框架在晶体结构预测和空间群预测任务上的准确率分别达到97.5%和98.7%.这一设计显著提升了高通量分析的准确性和效率,有望推动材料科学研究的进展,为解决其他具有高度关联性的多任务问题提供宝贵的启示.
High-throughput X-ray diffraction(XRD)analysis plays a crucial role in accelerating material discovery.However,traditional methods often rely heavily on manual interpretation and tend to overlook low-intensity peak information when processing complex XRD data,thus limiting the potential for accuracy improvement.To address this issue,a large language model(LLM)agent framework for multi-modal crystal structure prediction is proposed.The framework integrates a GPT-4-driven intelligent agent with multi-modal voting models based on XRD and pair distribution functions,enabling autonomous crystal structure and space group prediction tasks.In addition,the reasoning capability of the LLM is enhanced through the introduction of knowledge graphs,which aid in understanding the relationships between crystal features,thus improving both prediction accuracy and reasoning performance.Experimental results demonstrate that the accuracy of this framework in crystal structure prediction and space group prediction tasks reaches 97.5%and 98.7%,respectively.This design significantly enhances the accuracy and efficiency of high-throughput analysis,with the potential to advance materials science research and provide valuable insights for addressing other highly interrelated multi-task problems.
作者
曹芊
徐殷
肖明军
CAO Qian;XU Yin;XIAO Ming-Jun(School of Artificial Intelligence and Data Science,University of Science and Technology of China,Hefei 230026,China;School of Computer Science and Technology,University of Science and Technology of China,Hefei 230027,China;Suzhou Institute for Advanced Research,University of Science and Technology of China,Suzhou 215004,China)
出处
《计算机系统应用》
2025年第8期33-42,共10页
Computer Systems & Applications
基金
国家自然科学基金面上项目(62172386)
江苏省自然科学基金面上项目(BK20231212)。
关键词
大语言模型
深度学习
知识图谱
多模态融合
晶体结构预测
large language model(LLM)
deep learning
knowledge graph
multimodal fusion
crystal structure prediction
