知识卡片
交互范式按决策归属区分
外部程序先取资料再交模型回答是被动交互;固定流程规定工具调用次序是显式交互;由模型选择分解、工具和下一步则具有自主性。应说明具体哪些决策交给模型,而非仅用是否调用工具判断是不是智能体。
这张卡片看不懂?
这张卡片已通过人工审查,不再接受失败反馈。
已有反馈 · 0 条
还没有人提交反馈。
在整书笔记中的位置
- · 外部工具与智能体
原书出处 · 1 处
- · 查看原书依据
LLM Interaction Paradigms Suppose you have a task you want the LLM to solve. There are several possible options: The LLM uses its own memory and capabilities encoded in its parameters to solve the task. You feed the LLM all the context it needs to solve the task within the prompt, and the LLM uses the provided context and its capabilities to solve it. The
参考来源
- 位置:《Designing Large Language Model Applications》第10章《外部工具与智能体》(来源项目:a19c7933-4eda-4176-aebd-30fe7e1ad292;源文件:extracted/Designing Large Language Model Applications/_epub-src/OEBPS/ch10.html)
- 结论依据:摘录中关于交互范式按决策归属区分的定义、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:LLM Interaction Paradigms Suppose you have a task you want the LLM to solve. There are several possible options: The LLM uses its own memory and capabilities encoded in its parameters to solve the task. You feed the LLM all the context it needs to solve the task within the prompt, and the LLM uses the provided context and its capabilities to solve it. The