{"blog_url":"https://tech.revcomm.co.jp/","blog_title":"RevComm Tech Blog","title":"LLM-as-a-Judge\u3067\u300c\u3048\u3053\u3072\u3044\u304d\u300d\u304c\u8d77\u304d\u308b\uff1f\u2015\u2015 \u30bb\u30eb\u30d5\u30d0\u30a4\u30a2\u30b9\u3092\u7d71\u8a08\u7684\u306b\u6e2c\u308b\u624b\u6cd5\u306e\u7d39\u4ecb","categories":["\u751f\u6210AI","Research"],"provider_name":"Hatena Blog","version":"1.0","type":"rich","width":"100%","image_url":"https://cdn-ak.f.st-hatena.com/images/fotolife/r/revcomm-tech/20260624/20260624125257.png","url":"https://tech.revcomm.co.jp/measuring-self-bias-in-llm-as-a-judge","html":"<iframe src=\"https://hatenablog-parts.com/embed?url=https%3A%2F%2Ftech.revcomm.co.jp%2Fmeasuring-self-bias-in-llm-as-a-judge\" title=\"LLM-as-a-Judge\u3067\u300c\u3048\u3053\u3072\u3044\u304d\u300d\u304c\u8d77\u304d\u308b\uff1f\u2015\u2015 \u30bb\u30eb\u30d5\u30d0\u30a4\u30a2\u30b9\u3092\u7d71\u8a08\u7684\u306b\u6e2c\u308b\u624b\u6cd5\u306e\u7d39\u4ecb - RevComm Tech Blog\" class=\"embed-card embed-blogcard\" scrolling=\"no\" frameborder=\"0\" style=\"display: block; width: 100%; height: 190px; max-width: 500px; margin: 10px 0px;\"></iframe>","provider_url":"https://hatena.blog","description":"\u306f\u3058\u3081\u306b \u3053\u3093\u306b\u3061\u306f\u3002Research Engineer\u306e\u9ad9\u702c\u3067\u3059\u3002 \u6700\u8fd1\u306eAI\u958b\u767a\u3067\u306f\u3001\u30e2\u30c7\u30eb\u304c\u751f\u6210\u3057\u305f\u30c6\u30ad\u30b9\u30c8\u306e\u54c1\u8cea\u3092\u4eba\u9593\u304c\u624b\u52d5\u3067\u30c1\u30a7\u30c3\u30af\u3059\u308b\u4ee3\u308f\u308a\u306b\u3001\u5225\u306eLLM\uff08\u5927\u898f\u6a21\u8a00\u8a9e\u30e2\u30c7\u30eb\uff09\u306b\u63a1\u70b9\u3055\u305b\u308b\u300cLLM-as-a-Judge\u300d\u3068\u3044\u3046\u624b\u6cd5\u304c\u5e83\u304f\u4f7f\u308f\u308c\u308b\u3088\u3046\u306b\u306a\u3063\u3066\u3044\u307e\u3059\u3002 \u4ee5\u524d\u306e\u8a18\u4e8b\u3067\u3082\u3001LLM-as-a-Judge\u3092GPT-4\u3068ChatGPT\u3067\u5b9f\u969b\u306e\u30bf\u30b9\u30af\u306b\u9069\u7528\u3057\u3001\u300c\u5b9f\u52d9\u63a1\u7528\u524d\u306b\u7cbe\u5ea6\u691c\u8a3c\u3092\u3059\u308b\u3053\u3068\u300d\u306e\u91cd\u8981\u6027\u3092\u7d39\u4ecb\u3057\u307e\u3057\u305f\u3002 tech.revcomm.co.jp \u4eca\u56de\u306f\u305d\u306e\u4e00\u6b69\u5148\u306e\u554f\u3044\u3092\u6271\u3044\u307e\u3059\u3002 \u300c\u8a55\u4fa1\u8005\u306eLLM\u304c\u3001\u81ea\u5206\u81ea\u8eab\u306e\u56de\u7b54\u3092\u4e0d\u5f53\u306b\u9ad8\u304f\u8a55\u4fa1\u3057\u3066\u3057\u307e\u3046\u306e\u3067\u306f\u306a\u3044\u304b\uff1f\u300d \u5f93\u6765\u306e\u4eba\u624b\u8a55\u4fa1\u3068\u2026","height":"190","author_name":"yasaka_uta","author_url":"https://blog.hatena.ne.jp/yasaka_uta/","published":"2026-06-24 12:52:55"}