add evaluation-guidebook notes: HF guidebook knowledge (2024 GitHub + 2025 Space) into LLM_Evaluation

- Add 9 Chinese distilled notes under 04-Reference-Archive/evaluation-guidebook/
  covering automated benchmarks, human evaluation, LLM-as-a-judge,
  troubleshooting, general knowledge, yearly dives, resources, and the
  2025 HF Space edition (FineWeb eval-selection methodology, MCF/CF/FG,
  sampling metrics, Math-Verify, statistical validity & cost)
- Update zone README and Material-List with entry links
This commit is contained in:
windyboy
2026-08-21 16:19:12 +08:00
parent f245e3132b
commit f375cd6134
11 changed files with 3653 additions and 0 deletions
@@ -63,3 +63,4 @@ created: 2026-08-21
- 职业与岗位背景:[[02_Areas/Job/llm_data_annotation_programmer_roadmap|大模型数据标注与程序员入门]](原文存于 02_Areas/Job
- 入门认知草案:[[02-Intro-Cognitive-Framework-Draft|入门认知框架(草案)]]
- 路线图重构草案:[[03-Roadmap-Refactor-Outline-Draft|实战路线图重构大纲(草案)]]
- 外部权威知识参考:[[04-Reference-Archive/evaluation-guidebook/00-Overview|HuggingFace Evaluation Guidebook 中文提炼]](自动基准 / 人工评测 / LLM-as-judge / 排错,按主题查阅)