面向农业应用的通用大语言模型微调:基于Qwen3-8B的可复现框架与评估协议
Fine-Tuning General-Purpose Large Language Models for Agricultural Applications:A Reproducible Framework and Evaluation Protocol Based on Qwen3-8B
AI总结:
提出AgriTune-R框架,通过数据治理、指令构建、LoRA/QLoRA微调、检索增强生成和专家评估,将通用大语言模型适配到农业任务,并设计了包含事实性、安全性、证据一致性和不确定性表达的评估协议。
摘要:
通用大语言模型(LLM)在开放域问答、信息抽取和文本生成方面展现了强大能力。然而,农业应用具有领域特定、区域依赖、时间敏感和安全关键的特点。没有数据治理、专家评估和证据约束,农业助手可能在作物病害、农药使用、施肥或政策解读方面产生不可靠的建议。为避免将未经验证的模拟数字作为真实实验结果呈现,本文不报告任何未经实际训练运行和专家评估的模型性能声明。相反,我们提出了AgriTune-R,一个可复现且可审计的框架,用于将通用LLM适配到农业任务。该框架选择公开可验证的Qwen3-8B模型作为推荐基础模型,并整合了农业数据治理、指令构建、LoRA/QLoRA参数高效微调、检索增强生成、专家评估以及对高风险问题的安全控制。贡献包括:(1) 一个结构化的农业LLM适配工作流;(2) 一个针对农业知识问答、病虫害咨询、栽培管理和政策解读的评估协议;(3) 一个结合事实性、安全性、证据一致性和不确定性表达的专家评审标准;(4) 协议设计与实证结论的明确分离,为未来的实证研究提供了可执行的基线。
英文摘要:
General-purpose large language models (LLMs) have demonstrated strong abilities in opendomain question answering, information extraction, and text generation. Agricultural applications, however, are domain-specific, region-dependent, time-sensitive, and safety-critical. Without data governance, expert evaluation, and evidence constraints, an agricultural assistant mayproduce unreliable advice on crop diseases, pesticide use, fertilization, or policy interpretation.To avoid presenting unverified simulated numbers as real experimental findings, this paper doesnot report any model-performance claims that have not been produced by an actual training runand expert evaluation. Instead, we propose AgriTune-R, a reproducible and auditable frameworkfor adapting general-purpose LLMs to agricultural tasks. The framework selects the publiclyverifiable Qwen3-8B model as the recommended base model and integrates agricultural datagovernance, instruction construction, LoRA/QLoRA parameter-efficient fine-tuning, retrievalaugmented generation, expert evaluation, and safety control for high-risk questions. The contributions are: (1) a structured workflow for agricultural LLM adaptation; (2) an evaluationprotocol for agricultural knowledge QA, pest and disease consultation, cultivation management,and policy explanation; (3) an expert-review rubric combining factuality, safety, evidence consistency, and uncertainty expression; and (4) a clear separation between protocol design andempirical conclusions, providing an executable baseline for future empirical studies.
机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。