CyberSecEval

CyberSecEval The dataset source can be found here. (CyberSecEval2 Version) Abstract Large language models (LLMs) introduce new security risks, but there are few comprehensive evaluation suites to measure and reduce these risks. We present CYBERSECEVAL 2, a novel benchmark to quantify LLM security risks and capabilities

Papers using CyberSecEval (3)

CWEval: Outcome-driven Evaluation on Functionality and Security of LLM Code Generation2025 · 5 citesPurple Llama CyberSecEval: A Secure Coding Benchmark for Language Models2023 · 19 citesSecCodePLT: A Unified Platform for Evaluating the Security of Code GenAI2024