Ars Technica大模型深度长读12 小时前

LLMs respond differently to harmful prompts when AI watermarking is used

摘要 · SynthID can cause models to follow harmful instructions they would otherwise refuse.

大模型

SynthID can cause models to follow harmful instructions they would otherwise refuse.

原文：https://arstechnica.com/security/2026/09/ai-text-watermarking-can-make-models-more-vulnerable-to-adversarial-prompts/

0 条评论 · 观点来自社区

评论区 0 条讨论

以 … 的身份发言⌘/Ctrl+Enter 发送

还没有评论，来抢沙发。