{"provider_name":"Hatena Blog","image_url":"https://cdn-ak.f.st-hatena.com/images/fotolife/a/algo_kun/20260523/20260523172825.png","blog_url":"https://algo-kun.hatenablog.com/","categories":["Anthropic","Claude","Eval","AIAgent","NotebookLM"],"blog_title":"Anthropic / Claude \u6700\u65b0\u60c5\u5831\u307e\u3068\u3081","title":"AI \u30a8\u30fc\u30b8\u30a7\u30f3\u30c8\u306e\u4e3b\u89b3\u7684\u54c1\u8cea\u3092\u8a55\u4fa1\u3059\u308b\u300cEval\u300d\u5b9f\u8df5\u30ac\u30a4\u30c9","provider_url":"https://hatena.blog","published":"2026-05-23 21:01:40","author_url":"https://blog.hatena.ne.jp/algo_kun/","version":"1.0","height":"190","description":"AI \u30a8\u30fc\u30b8\u30a7\u30f3\u30c8\u306e\u4e3b\u89b3\u7684\u54c1\u8cea\u3092\u8a55\u4fa1\u3059\u308b\u300cEval\u300d\u5b9f\u8df5\u30ac\u30a4\u30c9 Anthropic \u306e\u30bb\u30c3\u30b7\u30e7\u30f3\u52d5\u753b Evals for taste: Hill-climbing a slide-generation agent \u306e\u5185\u5bb9\u3092\u65e5\u672c\u8a9e\u3067\u307e\u3068\u3081\u307e\u3059\u3002\u30b9\u30e9\u30a4\u30c9\u751f\u6210\u30a8\u30fc\u30b8\u30a7\u30f3\u30c8\u3092\u984c\u6750\u306b\u3001\u6c7a\u5b9a\u8ad6\u7684\u306a\u30b3\u30fc\u30c9\u8a55\u4fa1\u3068 LLM \u30d9\u30fc\u30b9\u306e Judge \u3092\u7d44\u307f\u5408\u308f\u305b\u305f\u54c1\u8cea\u6e2c\u5b9a\u3001\u81ea\u5df1\u56de\u5e30\u30e2\u30c7\u30eb\u306e\u843d\u3068\u3057\u7a74\u3001QA \u30eb\u30fc\u30d7\u306b\u3088\u308b\u7d99\u7d9a\u7684\u6539\u5584\u3001\u305d\u3057\u3066\u30e2\u30c7\u30eb\u5909\u66f4\u3060\u3051\u3067\u300ctaste\u300d\u3092\u5411\u4e0a\u3055\u305b\u308b\u4e8b\u4f8b\u307e\u3067\u3001\u5b9f\u52d9\u306b\u76f4\u7d50\u3059\u308b Eval \u8a2d\u8a08\u624b\u6cd5\u3092\u89e3\u8aac\u3057\u3066\u3044\u307e\u3059\u3002 \ud83d\udcdd \u4e0a\u8a18\u30a4\u30f3\u30d5\u30a9\u30b0\u30e9\u30d5\u30a3\u30c3\u30af\u306f Google NotebookLM \u306e Stu\u2026","type":"rich","author_name":"algo_kun","html":"<iframe src=\"https://hatenablog-parts.com/embed?url=https%3A%2F%2Falgo-kun.hatenablog.com%2Fentry%2F2026%2F05%2F23%2F210140\" title=\"AI \u30a8\u30fc\u30b8\u30a7\u30f3\u30c8\u306e\u4e3b\u89b3\u7684\u54c1\u8cea\u3092\u8a55\u4fa1\u3059\u308b\u300cEval\u300d\u5b9f\u8df5\u30ac\u30a4\u30c9 - Anthropic / Claude \u6700\u65b0\u60c5\u5831\u307e\u3068\u3081\" class=\"embed-card embed-blogcard\" scrolling=\"no\" frameborder=\"0\" style=\"display: block; width: 100%; height: 190px; max-width: 500px; margin: 10px 0px;\"></iframe>","width":"100%","url":"https://algo-kun.hatenablog.com/entry/2026/05/23/210140"}