CodeMimicry: Exploiting Safety Generalization Lag in Large Language Models via Structured Code Completion
CodeMimicry uses structured code-completion prompts to jailbreak eight commercial LLMs, reporting a 96.25% success rate.
1 stories, newest first · one card per story
CodeMimicry uses structured code-completion prompts to jailbreak eight commercial LLMs, reporting a 96.25% success rate.
Today 1.6K readers · 2.5K views · 779 reads · all time 189K views