知识卡片

普通读书笔记卡 · 6644.b.3 · 生成模型:GPT-6

会话记忆与通用知识分开存取

理解等级:中级

数据库可保存历史提及、回应、用户信息与领域知识,分别支持连续对话、个性化和事实检索。它们更新频率与可信程度不同,不宜把上一轮生成文本直接升级为世界事实;有些简单机器人也不需要完整知识库。

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

      Database —Store conversation history, user information, and general world knowledge for use in the Decide subsystem. Each of these four stages can be implemented using one or more of the algorithms listed within the corresponding boxes in the block diagram. And you can combine the Python examples from this book in your own NLP system to accomplish state-of-the-art performance for most applications. By the end

    参考来源

    - 位置:《Natural Language Processing in Action, Second Edition》第01章《会读写的机器与自然语言处理》(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-1.html) - 结论依据:摘录中关于会话记忆与通用知识分开存取的定义、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:Database —Store conversation history, user information, and general world knowledge for use in the Decide subsystem. Each of these four stages can be implemented using one or more of the algorithms listed within the corresponding boxes in the block diagram. And you can combine the Python examples from this book in your own NLP system to accomplish state-of-the-art performance for most applications. By the end