知识卡片
固定工作流约束步骤自由度
预设先取数据、再分析、再可视化等步骤,可以减少模型自由决定执行路径的范围。固定次序有助于定位和验证故障,但仍须检查每一步参数、数据和结果,不能把确定流程等同确定正确。
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
The Explicit Approach Figure 10-2 demonstrates the explicit approach to interface LLMs with external tools. Figure 10-2. The explicit interaction approach in action Unlike in the passive approach, the LLM is no longer a passive participant. We provide the LLM with explicit instructions on how and when to invoke external data stores and tools. The LLM interacts with its environment based on a pre-programmed set
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:摘录中关于固定工作流约束步骤自由度的定义、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:The Explicit Approach Figure 10-2 demonstrates the explicit approach to interface LLMs with external tools. Figure 10-2. The explicit interaction approach in action Unlike in the passive approach, the LLM is no longer a passive participant. We provide the LLM with explicit instructions on how and when to invoke external data stores and tools. The LLM interacts with its environment based on a pre-programmed set