https://hongsymbol.online/tags/eval/Recent content in Eval on Symbol's NotesHugozh-cnMon, 11 May 2026 19:00:00 +0800https://hongsymbol.online/posts/agent-eval/Mon, 11 May 2026 19:00:00 +0800https://hongsymbol.online/posts/agent-eval/改了 Prompt 感觉没问题→上线→用户反馈变蠢了→永远在救火。从 Vibe Check 进化到 EDD（评测驱动开发），Anthropic 工程博客精华提炼——三种评分器、两套评测套件、pass@k vs pass^k 核心指标，8步从零搭建评测体系。