AI RESEARCH
Gradient-Controlled Decoding: A Safety Guardrail for LLMs with Dual-Anchor Steering
arXiv CS.CL
•
ArXi:2604.05179v1 Announce Type: new Large language models (LLMs) remain susceptible to jailbreak and direct prompt-injection attacks, yet the strongest defensive filters frequently over-refuse benign queries and degrade user experience. Previous work on jailbreak and prompt injection detection such as GradSafe, detects unsafe prompts with a single "accept all" anchor token, but its threshold is brittle and it offers no deterministic guarantee that harmful content will not be emitted once decoding begins. We.