cyberseceval
LLM security
AI security
CyberSecEval 2 for LLM security assessments
Explore CyberSecEval 2, a framework assessing security risks in large language models like GPT-4 and Meta Llama 3, highlighting vulnerabilities, safety-utility trade-offs, and practical applications for enhancing AI security.
31 May 2024 ยท 16 min read
