Lilian Weng·· 2024-12-02精选
🦃 At the end of Thanksgiving holidays, I finally finished the piece on reward hacking. Not an easy ...
🦃 At the end of Thanksgiving holidays, I finally finished the piece on reward hacking. Not an easy ...
AI 导读
🦃 感恩节假期结束时,我终于完成了关于 reward hacking 的文章。不好写啊,呼。
推荐理由
OpenAI研究员深度解析奖励作弊机制,揭示LLM自主化部署的关键安全障碍
来源:Lilian Weng · x.com