arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2512.06404cs.AIcond-mat.mtrl-sciphysics.chem-ph
收藏

GENIUS: 一种用于自主设计和执行模拟协议的智能AI框架

GENIUS: An Agentic AI Framework for Autonomous Design and Execution of Simulation Protocols

Karlsruhe Institute of Technology(卡尔斯鲁厄理工学院);Institute of Nanotechnology(纳米技术研究所);Hamburg University of Technology(汉堡技术大学);Federal University of Paraná(帕拉纳联邦大学);University of Brasília(巴西利亚大学);Federal University of Pelotas(普拉多斯联邦大学);Institute of Physics and International Center of Physics(物理研究所和国际物理中心)

Mohammad Soleymanibrojeni, Roland Aydin, Diego Guedes-Sobrinho, Alexandre C. Dias, Maurício J. Piotrowski, Wolfgang Wenzel, Celso Ricardo Caldeira Rêgo

AI总结:

提出GENIUS框架,融合量子ESPRESSO知识图谱与分层大语言模型,通过有限状态错误恢复机器实现从自由文本到验证输入文件的自动生成,在295个基准测试中达到约80%的成功率,并大幅降低推理成本与幻觉。

摘要:

预测性原子模拟推动了材料发现,但常规设置和调试仍需计算机专家。这种知识差距限制了集成计算材料工程(ICME),因为最先进的代码存在但非专家使用起来仍然繁琐。我们通过GENIUS解决了这一瓶颈,这是一种AI智能体工作流,将智能的Quantum ESPRESSO知识图谱与由有限状态错误恢复机器监督的分层大语言模型层次结构融合。我们展示了GENIUS将自由形式的人类生成提示翻译成经过验证的输入文件,在295个多样化基准测试中约有80%运行完成,其中76%被自主修复,成功率呈指数衰减至7%的基线。与仅使用LLM的基线相比,GENIUS将推理成本减半,并几乎消除了幻觉。该框架通过智能自动化协议生成、验证和修复,使电子结构DFT模拟大众化,为全球学术界和工业界开放大规模筛选并加速ICME设计循环。

英文摘要:

Predictive atomistic simulations have propelled materials discovery, yet routine setup and debugging still demand computer specialists. This know-how gap limits Integrated Computational Materials Engineering (ICME), where state-of-the-art codes exist but remain cumbersome for non-experts. We address this bottleneck with GENIUS, an AI-agentic workflow that fuses a smart Quantum ESPRESSO knowledge graph with a tiered hierarchy of large language models supervised by a finite-state error-recovery machine. Here we show that GENIUS translates free-form human-generated prompts into validated input files that run to completion on $\approx$80% of 295 diverse benchmarks, where 76% are autonomously repaired, with success decaying exponentially to a 7% baseline. Compared with LLM-only baselines, GENIUS halves inference costs and virtually eliminates hallucinations. The framework democratizes electronic-structure DFT simulations by intelligently automating protocol generation, validation, and repair, opening large-scale screening and accelerating ICME design loops across academia and industry worldwide.

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。

↑