domain-camouflage

This paper identifies a new class of injection attacks where payloads mimic the domain language to evade LLM injection detectors, showing detection rates drop dramatically (e.g., from 93.8% to 9.7% on Llama 3.1 8B). The vulnerability is systematic and extends to dedicated safety classifiers like Llama Guard 3, which detected zero camouflage payloads.