大语言模型在植物病虫害智能诊断中应用的研究进展OA

Plant diseases and pests are major constraints on agricultural production.Traditional diagnos-tic methods relying on manual visual inspection and expert experience are often characterized by low efficiency and inconsistent standards.Large language models(LLMs),with their powerful capabilities in language understanding,knowledge integration,and reasoning,offer new approaches for developing intelligent diagnostic systems for plant diseases and insect pests.This paper systematically reviews the technical foundations and recent developments of large language models,retrieval-augmented genera-tion(RAG)-based diagnostic methods,applications of multimodal large language models,agent-based diagnostic system architectures,as well as prompt engineering and model fine-tuning techniques.The advantages and limitations of existing studies are analyzed,and future research directions are discussed,with the aim of providing a reference for related research and practical applications.