知识卡片

普通读书笔记卡 · 2931.e.3.a.1.f.1

RAG改变输入而非模型参数

理解等级:中级

检索增强生成先按请求和租户语境寻找相关资料,再把它放入提示上下文交给模型。它改变本次输入,不直接训练或更新模型权重;检索可使用关系库、搜索或向量数据库,技术选择应服务资料与查询需求。

这张卡片看不懂?

已有反馈 · 0 / 10 条

    还没有人提交反馈。

    在整书笔记中的位置

    原书出处 · 1 处

    • GenAI and Multi-Tenancy · 查看原书依据

      GenAI service. Neither of these are true with RAG, which is exclusively focused on augmenting the prompt that gets sent into the GenAI service. This means its impact is achieved purely through changing or refining the nature of the request.

    参考来源

    - 位置:《Building Multi-Tenant SaaS Architectures》第16章《GenAI and Multi-Tenancy》(来源项目 a19c7933-4eda-4176-aebd-30fe7e1ad292;源文件:extracted/Building Multi-Tenant SaaS Architectures/_epub-src/OEBPS/ch16.html) - 结论依据:原文具体说明检索增强生成先按请求和租户语境寻找相关资料,再把它放入提示上下文交给模型;本卡据此归纳机制及适用边界。 - 原始内容:GenAI service. Neither of these are true with RAG, which is exclusively focused on augmenting the prompt that gets sent into the GenAI service. This means its impact is achieved purely through changing or refining the nature of the request.