DeepMind CaMeL论文防御提示注入
"Deterministic code checks the result" sounds like they might be implementing a variant of the DeepM...
DeepMind分享防御提示注入的实用方法，包含模型训练和代码检查多层防护。
DeepMind的CaMeL论文提出了一种防御提示注入的多层方法。模型经过训练以识别和抵抗提示注入。系统会标记来自不可信来源的内容。确定性代码会检查结果。分类器集合在代理无法访问的位置运行。
"Deterministic code checks the result" sounds like they might be implementing a variant of the DeepM...
"Deterministic code checks the result" sounds like they might be implementing a variant of the DeepMind CaMeL paper simonwillison.net/2025/Apr/11/ca… David Singleton @dps One threat we’re particularly focused on is prompt injection, and we handle it in layers. The model is trained to recognize and resist it. The harness marks anything coming from an untrusted source. Deterministic code checks the result. And an ensemble of classifiers runs where the agent can't reach them. 🔗 View Quoted Tweet 💬 3 🔄 1 ❤️ 8 👀 3246 📊 4 ⚡
