04:00
2026-08-28
arxiv.org
artificial-intelligence
NeuronFuzz: Safety Neuron Guided Fuzzing for LLM Safety Evaluation
NeuronFuzz, a white-box fuzzing framework introduced in an arXiv paper (2608.26222v1), uses internal safety neurons as continuous feedback to evaluate LLM safety, achieving a 76-100% jailbreak discove…