Saved in:
| Main Authors: | , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2512.19458 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866914214463930368 |
|---|---|
| author | Xia, Zeyu Ma, Jinzhe Zheng, Congjie Zhang, Shufei Li, Yuqiang Su, Hang Hu, P. Zhang, Changshui Gong, Xingao Ouyang, Wanli Bai, Lei Zhou, Dongzhan Su, Mao |
| author_facet | Xia, Zeyu Ma, Jinzhe Zheng, Congjie Zhang, Shufei Li, Yuqiang Su, Hang Hu, P. Zhang, Changshui Gong, Xingao Ouyang, Wanli Bai, Lei Zhou, Dongzhan Su, Mao |
| contents | Large Language Models (LLMs) have emerged as powerful tools for accelerating scientific discovery, yet their static knowledge and hallucination issues hinder autonomous research applications. Recent advances integrate LLMs into agentic frameworks, enabling retrieval, reasoning, and tool use for complex scientific workflows. Here, we present a domain-specialized agent designed for reliable automation of first-principles materials computations. By embedding domain expertise, the agent ensures physically coherent multi-step workflows and consistently selects convergent, well-posed parameters, thereby enabling reliable end-to-end computational execution. A new benchmark of diverse computational tasks demonstrates that our system significantly outperforms standalone LLMs in both accuracy and robustness. This work establishes a verifiable foundation for autonomous computational experimentation and represents a key step toward fully automated scientific discovery. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2512_19458 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | An Agentic Framework for Autonomous Materials Computation Xia, Zeyu Ma, Jinzhe Zheng, Congjie Zhang, Shufei Li, Yuqiang Su, Hang Hu, P. Zhang, Changshui Gong, Xingao Ouyang, Wanli Bai, Lei Zhou, Dongzhan Su, Mao Artificial Intelligence Materials Science Large Language Models (LLMs) have emerged as powerful tools for accelerating scientific discovery, yet their static knowledge and hallucination issues hinder autonomous research applications. Recent advances integrate LLMs into agentic frameworks, enabling retrieval, reasoning, and tool use for complex scientific workflows. Here, we present a domain-specialized agent designed for reliable automation of first-principles materials computations. By embedding domain expertise, the agent ensures physically coherent multi-step workflows and consistently selects convergent, well-posed parameters, thereby enabling reliable end-to-end computational execution. A new benchmark of diverse computational tasks demonstrates that our system significantly outperforms standalone LLMs in both accuracy and robustness. This work establishes a verifiable foundation for autonomous computational experimentation and represents a key step toward fully automated scientific discovery. |
| title | An Agentic Framework for Autonomous Materials Computation |
| topic | Artificial Intelligence Materials Science |
| url | https://arxiv.org/abs/2512.19458 |