中国铁路2024,Issue(7) :7-14.DOI:10.19549/j.issn.1001-683x.2024.03.08.001

铁路自然语言大模型关键技术研究及应用展望

Key Technologies and Application Prospects of Railway Natural Language Large Model

史天运 李新琴 代明睿 史维峰 李国华 杜文然
中国铁路2024,Issue(7) :7-14.DOI:10.19549/j.issn.1001-683x.2024.03.08.001

铁路自然语言大模型关键技术研究及应用展望

Key Technologies and Application Prospects of Railway Natural Language Large Model

史天运 1李新琴 2代明睿 2史维峰 2李国华 2杜文然2
扫码查看

作者信息

  • 1. 中国铁道科学研究院集团有限公司,北京 100081
  • 2. 中国铁道科学研究院集团有限公司电子计算技术研究所,北京 100081
  • 折叠

摘要

人工智能自然语言大模型的涌现为行业深度赋能带来了新的曙光,研究铁路自然语言大模型关键技术及应用,对促进和统筹铁路人工智能发展具有重要意义.根据铁路人工智能应用需求,提出铁路自然语言大模型应用场景;依托铁路人工智能平台,设计铁路自然语言大模型的总体架构,研究自然语言大模型关键技术,构建面向智能问答的铁路行业大模型,并以实际数据进行模型验证;最后从铁路运输组织、铁路运营安全、旅客服务方面对铁路自然语言大模型的发展和应用进行展望.

Abstract

The emergence of artificial intelligence natural language large models has brought new dawn for the in-depth empowerment of the industry.Research on key technologies and applications of railway natural language large models is of great significance to promoting and coordinating the development of railway artificial intelligence.This paper puts forward the application scenario of railway natural language large model according to the application requirements of railway artificial intelligence;designs the overall architecture of the railway natural language large model by relying on the railway artificial intelligence platform,studies the key technologies of the natural language large model,builds a railway industry large model oriented to intelligent Q&A,and verifies the model with actual data;finally,this paper prospects for the development and application of railway natural language large model from railway traffic organization,railway operation safety and passenger service.

关键词

智能高铁/人工智能/铁路自然语言大模型/应用场景/大模型架构/大模型微调/检索增强生成/铁路知识问答

Key words

intelligent HSR/artificial intelligence/railway natural language large model/application scenarios/large model architecture/large model fine-tuning/retrieval-augmented generation/railway knowledge Q&A

引用本文复制引用

基金项目

中国国家铁路集团有限公司科技研究开发计划项目(P2023S001)

出版年

2024
中国铁路
中国铁道科学研究院

中国铁路

CSTPCD北大核心
影响因子:0.407
ISSN:1001-683X
参考文献量10
段落导航相关论文