知识卡片

普通读书笔记卡 · 6768.e · 生成模型:GPT-6

一次拒答成功不证明全面不编造

理解等级:中级

模型面对知识库外的卢浮宫问题答不知道,是一个测试结果。需覆盖可答、不可答、冲突、过期及检索误召回等情况,分别查漏答和无依据回答;不能由一条展示就宣布不会生成资料外内容。

这张卡片看不懂?

已有反馈 · 0 条

    还没有人提交反馈。

    在整书笔记中的位置

    原书出处 · 1 处

    • · 查看原书依据

      Well, that was a bit vague—but basically correct! And that’s a result you can expect for a relatively small model. Let’s make sure our model doesn’t make answers up: >>> question = "Where is the Louvre Museum?" >>> result = rag_pipe.run( query=question, ... params={"Retriever": {"top_k": 5}}) #1 >>> print(result['results'][0]) 'I don't know' #1 top_k is the number of documents the retriever fetches. This is exactly

    参考来源

    - 位置:《Natural Language Processing in Action, Second Edition》第10章《真实世界中的大语言模型》(来源项目:a19c7933-4eda-4176-aebd-30fe7e1ad292;源文件:extracted/Natural Language Processing in Action, Second Edition/_epub-src/OEBPS/Text/chapter-10.html) - 结论依据:摘录中关于一次拒答成功不证明全面不编造的定义、比较或条件支撑卡片;原书表述不严谨的范围已在覆盖记录逐项限定,不把案例结果外推为普遍结论。 - 原始内容:Well, that was a bit vague—but basically correct! And that’s a result you can expect for a relatively small model. Let’s make sure our model doesn’t make answers up: >>> question = "Where is the Louvre Museum?" >>> result = rag_pipe.run( query=question, ... params={"Retriever": {"top_k": 5}}) #1 >>> print(result['results'][0]) 'I don't know' #1 top_k is the number of documents the retriever fetches. This is exactly