← 返回刚才阅读的位置

知识卡片

普通读书笔记卡 · 6642.e · 生成模型:GPT-6

多数投票聚合答案并非验证真值

理解等级:中级

自一致性产生多条解题轨迹并归并终值投票,依赖答案等价规则、无效票和平票处理。共同知识缺口会让多次结果一致地出错,因此投票只是选择策略,需用独立任务标准测其收益与成本。

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

      Self-consistency is a popular prompting technique that uses majority voting in conjunction with CoT prompting. In this technique, we add the CoT prompt “Let’s think step by step” to the input and run multiple generations (reasoning paths). We then use majority voting to select the correct output.

    参考来源

    - 位置:《Designing Large Language Model Applications》第05章《按任务适配模型》(来源项目:a19c7933-4eda-4176-aebd-30fe7e1ad292;源文件:extracted/Designing Large Language Model Applications/_epub-src/OEBPS/ch05.html) - 结论依据:摘录中关于多数投票聚合答案并非验证真值的定义、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:Self-consistency is a popular prompting technique that uses majority voting in conjunction with CoT prompting. In this technique, we add the CoT prompt “Let’s think step by step” to the input and run multiple generations (reasoning paths). We then use majority voting to select the correct output.