LLM记忆门控防御论文：评估激活与记忆诱导谄媚

1. 12:00
   arXiv · Artificial Intelligence

   LLM 防御纵深：评估记忆门控对抗激活诱导与记忆诱导谄媚

   研究提出 2×2 防御纵深框架，分离内部激活引导与外部记忆处理，在 MemSyco-Bench 上评测四个开源模型、10 个引导系数与五种记忆防御配置。Selective Router Gate 比完全移除记忆保留更多平均准确率，且在模型被引导偏向谄媚时仍成立。

Current heat 9·Comparable peak 10（Oct 7）·Comparable change over 24 hours –

The trend compares only the same participants observed continuously; its range may be smaller than the current heat count. Move or click on the chart to inspect hourly heat; use the left and right arrow keys to switch.