← 返回刚才阅读的位置

知识卡片

普通读书笔记卡 · 6517.e.1 · 生成模型:GPT-6

裁判能力也要人工校准

理解等级:中级

裁判模型、提示和风格偏好会改变评分。应对代表性好坏样本核对人与裁判的分歧、检查对无关改写的敏感性,再决定其使用范围;可扩展的自动评分不等于独立客观事实。

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

      large pools of human voters. (Technically, it is possible to outsource the preference-based rating behind leaderboards to LLM judges as well). However, LLM judges also share similar weaknesses with human voters: results can be biased by model preferences, prompt design, and answer style. Also, there is a strong dependency on the choice of judge model and rubric, and they lack the reproducibility of fixed benchmarks.

    参考来源

    - 位置:《Build a Reasoning Model (From Scratch)》第10章《附录F 常见模型评估方法》(来源项目:a19c7933-4eda-4176-aebd-30fe7e1ad292;源文件:extracted/Build a Reasoning Model (From Scratch)/_epub-src/OEBPS/Text/appendix-f.html) - 结论依据:摘录中关于裁判能力也要人工校准的定义、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:large pools of human voters. (Technically, it is possible to outsource the preference-based rating behind leaderboards to LLM judges as well). However, LLM judges also share similar weaknesses with human voters: results can be biased by model preferences, prompt design, and answer style. Also, there is a strong dependency on the choice of judge model and rubric, and they lack the reproducibility of fixed benchmarks.