知识卡片
评估成本包括裁判调用
用模型替代逐题人类投票可加快批量评估,但需计算裁判生成、参考准备、重复评分及人工校准的成本。更强裁判也可能具有更多资源要求;方案比较应把这些额外调用与待测模型推理分开。
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
原书出处 · 1 处
- · 查看原书依据
do not rely on 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
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:摘录中关于评估成本包括裁判调用的定义、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:do not rely on 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