vault backup: 2026-09-16 17:56:55
This commit is contained in:
@@ -16,7 +16,9 @@ created: 2026-08-21
|
||||
1. [[01-What-Is-LLM-Evaluation|什么是 LLM Evaluation]]
|
||||
2. [[02-Annotation-Human-Data-and-Evaluation|数据标注、Human Data 与 Evaluation]]
|
||||
3. [[03-Core-Concept-Map|LLM Evaluation 核心概念地图]]
|
||||
4. [[04-Concepts-and-Theory|LLM 评测基础概念与理论:系统讲解]](前三篇的教学整合:构念→测量→决策主线、信度/效度、pass@k 与 pass^k、capability/regression eval、统计不确定性与失败分析等)
|
||||
4. [[04-Concepts-and-Theory|LLM 评测基础概念与理论:系统讲解]](**可选深读,非首周必读**;前三篇的教学整合:构念→测量→决策主线、信度/效度、pass@k 与 pass^k、capability/regression eval、统计不确定性与失败分析等)
|
||||
|
||||
> 首周有界理论 = 01–03 三篇 + [[02-Why-Guide]];`04-Concepts-and-Theory` 按章节按需查阅,通读会超出 3–4 小时预算。
|
||||
|
||||
完成这一层后,继续阅读:
|
||||
|
||||
@@ -26,3 +28,6 @@ created: 2026-08-21
|
||||
> ⚠️ **读完本层 ≠ 完成理论学习。** 理论阶段的出口是 [[01-Learning-Board|学习看板]] Level 1 全部勾选 + [[02-First-Week-Worksheet|首周工作表]] 第 0 节填完;本层只负责建立共同语言。读完就进入 Why 与工作表,而不是继续读 [[01_Projects/Personal-Tech/LLM_Evaluation/04-Reference/README|04-Reference]]。
|
||||
|
||||
原则:这里回答 **What**,后续文档分别回答 **Why** 和 **How**。
|
||||
|
||||
|
||||
[You have received this identical output 3 times. Re-reading '/Users/windy/Documents/vault/my-vault/01_Projects/Personal-Tech/LLM_Evaluation/00-Foundations/README.md:raw' will not change it — use a narrower selector (path:A-B), or proceed with the edit.]
|
||||
Reference in New Issue
Block a user