← 返回起始笔记位置

知识卡片

普通读书笔记卡 · 8674 · 生成模型:GPT-6

对齐目标需要拆解为可评估行为

理解等级:中级

帮助性要求有效完成合理请求,诚实性要求准确并表达不确定性,无害性要求避免有害输出。三者不是一个总分:拒绝过多可损害帮助性,讨好用户可损害诚实性;需明确情境、标注规范和冲突处理。

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

      Helpful As long as a user request isn’t harmful, the AI should attempt to solve the request as effectively as possible, asking follow-up questions if needed. Honest The AI should provide accurate information and should be calibrated, providing reasonably accurate uncertainty estimates. It should understand its shortcomings. Harmless The AI should not be offensive or discriminatory and should refuse to perform tasks that can cause

    参考来源

    - 位置:《Designing Large Language Model Applications》第08章《对齐训练与推理》(来源项目:a19c7933-4eda-4176-aebd-30fe7e1ad292;源文件:extracted/Designing Large Language Model Applications/_epub-src/OEBPS/ch08.html) - 结论依据:摘录中关于对齐目标需要拆解为可评估行为的定义、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:Helpful As long as a user request isn’t harmful, the AI should attempt to solve the request as effectively as possible, asking follow-up questions if needed. Honest The AI should provide accurate information and should be calibrated, providing reasonably accurate uncertainty estimates. It should understand its shortcomings. Harmless The AI should not be offensive or discriminatory and should refuse to perform tasks that can cause