跳至内容

拾光小记

标签: Deceptive Behaviors

The MASK Benchmark: Disentangling Honesty From Accuracy in AI Systems

论文标题:The MASK Benchmark: Disentangling Honesty From Accuracy in AI Systems 原文链接:http://arxiv.org/pdf/2503.03750v1 发表时间:2025-03-05 18:59:23 作者:Richard …