知识卡片
重排分数要核对激活函数
num_labels=1只说明单输出维度,模型保存配置或调用参数还决定是否应用sigmoid。原始logit、sigmoid分数和真实相关概率不同;即使范围落在0到1,也需校准才能解释为概率。
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
- · 检索增强生成
原书出处 · 1 处
- · 查看原书依据
Because we have set num_labels = 1, the model will treat it as a regression task, using the sigmoid activation function to output a score between 0 and 1. These days, more advanced models like Contextualized Late Interaction over BERT (ColBERT) are used for reranking. As opposed to the cross-encoder setup we just discussed, ColBERT-style models allow for pre-computation of document representations, leading to faster
参考来源
- 位置:《Designing Large Language Model Applications》第12章《检索增强生成》(来源项目:a19c7933-4eda-4176-aebd-30fe7e1ad292;源文件:extracted/Designing Large Language Model Applications/_epub-src/OEBPS/ch12.html)
- 结论依据:摘录中关于重排分数要核对激活函数的定义、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:Because we have set num_labels = 1, the model will treat it as a regression task, using the sigmoid activation function to output a score between 0 and 1. These days, more advanced models like Contextualized Late Interaction over BERT (ColBERT) are used for reranking. As opposed to the cross-encoder setup we just discussed, ColBERT-style models allow for pre-computation of document representations, leading to faster