TopoChat:Enhancing Topological Materials Retrieval with Large Language Model and Multi-Source KnowledgeOA
Large language models(LLMs)perform well in general text tasks but face challenges in specialized fields like materials science.We present TopoChat,a knowledge-enhanced question-answering framework for materials science,which combines a domain-specific knowledge graph(TopoKG,Topological Materials Knowledge Graph)and a literature clustering module.TopoChat retrieves both relevant subgraphs and literature information for each query,integrating structured and unstructured knowledge to support LLM reasoning.Experiments on two benchmarks,MaScQA and TopoQA,show that TopoChat improves answer accuracy across multiple LLMs.These results demonstrate that integrating knowledge graphs and literature context enhances reliability in scientific question answering.TopoChat provides an effective approach for adapting LLMs to complex domains,narrowing the gap between general language abilities and domain expertise.
Huang-Chao Xu;Bao-Hua Zhang;Zhong Jin;Tian-Nian Zhu;Quan-Sheng Wu;Hong-Ming Weng
Computer Network Information Center,Chinese Academy of Sciences,Beijing 100083,China University of Chinese Academy of Sciences,Beijing 101408,ChinaComputer Network Information Center,Chinese Academy of Sciences,Beijing 100083,ChinaComputer Network Information Center,Chinese Academy of Sciences,Beijing 100083,ChinaUniversity of Chinese Academy of Sciences,Beijing 101408,China Institute of Physics,Chinese Academy of Sciences,Beijing 100190,ChinaInstitute of Physics,Chinese Academy of Sciences,Beijing 100190,ChinaInstitute of Physics,Chinese Academy of Sciences,Beijing 100190,China
信息技术与安全科学
information retrievallarge language model(LLM)knowledge graphprompt learning
《Journal of Computer Science & Technology》 2026 (2)
P.684-697,14
supported by the Informatization Plan of Chinese Academy of Sciences under Grant No.CAS-WX2021SF-0102.
评论