SOURCE-LINKED INTELLIGENCE

Semantic Overlays: Mitigating Prompt Injection with Annotations Beyond Tokens and Steering Vectors

arXiv · AI, language, vision and robotics · article · Aug 24, 2026 · UTC

Everything a language model sees is tokens. The serving stack knows what each span is -- user input, tool output, instructions -- but the model must keep track of that itself, and can lose track or be confused: text can be written to read like anything. Prompt injection is a natural exploit of this phenomenon. By scrambling the model's understanding of span identity, an attacker can induce unwanted and dangerous actions. Adding a non-textual channel to the model's input -- a way to communicate span identity beyond text -- mitigates this class of attack. We thus introduce a general steering tec

Read original source ↗ Open in workspace

recordType

paper

region

Global