返回探索

Jev 的结构化输出为何能抵御提示注入

帖子指出,提示注入之所以有效,是因为模型能写出下一条指令;而 Jev 的结构化评估模型只输出类型化的概率、标签或刻度值,恶意文本只能影响数值,无法变成命令。

Prompt injection works because the model can write the next instruction. A structured evaluation model cannot. Jev's only output is typed a probability, a label, a point on a scale. Text that tries to hijack an agent can move the number. It cannot become a command. That is why

Jev 的结构化输出为何能抵御提示注入 1
· 0 次赞在 X 打开