https://fisherdaddy.com/tags/ai%E5%AE%89%E5%85%A8/ Recent content in AI安全 on FisherAI https://fisherdaddy.com/images/papermod-cover.png https://fisherdaddy.com/images/papermod-cover.png Hugo -- 0.125.7 en Tue, 09 Jun 2026 18:59:00 +0800 https://fisherdaddy.com/posts/implications-large-scale-test-time-compute/ Tue, 09 Jun 2026 18:59:00 +0800 https://fisherdaddy.com/posts/implications-large-scale-test-time-compute/ Noam Brown 讨论大模型评估为何必须纳入测试时计算量，以及这对能力测量和 AI 安全准备工作的影响。