期刊文献+

GPT技术驱动的农业发展范式研究与展望

Research and Prospects on the Agricultural Development Paradigm Driven by GPT Technology
下载PDF
导出
摘要 当前,机器学习、自然语言理解等人工智能技术持续演进,Transformer模型架构、基于人工反馈的强化学习等关键技术取得重要突破。GPT(Generative Pre-trained Transformer)技术驱动的生成式人工智能大语言模型开发取得显著进展,创新生态有寒武纪大爆发的趋势,正在加快通用人工智能时代的到来。GPT技术与农业农村深入渗透融合,将引发农业农村领域智能化变革,对农业农村产生广泛而深远的影响。农业人工智能多模态大模型将成为农业领域科技创新的热点。本研究分析了人工智能技术在智能育种、作物和土壤智能监测预警、病虫草害智能识别预警、农业机器人、智能专家系统等领域的应用,提出了构建大规模农业标注语料库、研建农业多模态大模型、开发多模态大模型智能交互终端、部署农业大模型场景应用等农业GPT技术开发路径,对GPT技术促进农业生产、经营、管理、服务、科研等领域的范式创新进行了分析和展望。 Currently,artificial intelligence technologies such as machine learning and natural language understanding are continuously evolving,with key advancements in the Transformer model architecture and reinforcement learning based on human feedback.The development of generative AI large language models driven by GPT(Generative Pre-trained Transformer)technology has made significant progress,indicating an innovation explosion reminiscent of the Cambrian explosion,and is accelerating the arrival of the era of general artificial intelligence.The deep integration of GPT technology with agriculture and rural areas is set to trigger a transformation towards intelligent operations and will have a wide-reaching and profound impact on these sectors.Multimodal large models in agricultural artificial intelligence are poised to become a hotbed of scientific and technological innovation within the field of agriculture.The application of AI technologies in smart breeding,intelligent crop and soil monitoring and early warning systems,intelligent identification and alert systems for pests and diseases,agricultural robots,and intelligent expert systems were analyzed.This paper suggested paths for the development of agricultural GPT technology,including the construction of large-scale agricultural annotated corpus,the research and construction of multimodal large models for agriculture,the development of intelligent multimodal model interaction terminals,and the deployment of large model-based agricultural application scenarios.The analysis also looked forward to the role of GPT technology in facilitating paradigmatic innovations in agricultural production,operations,management,services,and scientific research.
作者 李灯华 李干琼 许世卫 陈威 Li Denghua;Li Ganqiong;Xu Shiwei;Chen Wei(A gricultural Information Research Institute of Chinese Academy of Agricultural Sciences,Key Laboratory of Agricultural Monitoring and Early Warning Technology of Ministry of Agriculture and Rural Affairs,Beijing Agricultural Monitoring and Early Warning Engineering Technology Research Center,Beijing 100081)
出处 《农业展望》 2023年第12期73-80,共8页 Agricultural Outlook
基金 中国农业科学院科技创新工程项目(CAAS-ASTIP-2023-AII)。
关键词 GPT 人工智能 大语言模型 深度学习 农业 GPT artificial intelligence large language model deep learning agriculture
  • 相关文献

参考文献11

二级参考文献163

共引文献191

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部