ended4월 14일· 1 sources
Frontier LLMs Put to the Test in Real-World Vulnerability Discovery
LLM, 실제 취약점 찾아낼 수 있을까? N-Day-Bench 실전 보안 성능 검증
Why it matters
N-Day-Bench evaluates how frontier LLMs identify real-world vulnerabilities disclosed after their training cutoff, moving beyond synthetic tests to actual security auditing. This benchmark provides a critical assessment of AI's potential in autonomous cyber defense and the evolving landscape of AI-driven security research.
1
Sources
+0
24h
—
Growth
160d
Active
N-Day-BenchLLMvulnerability discoveryCybersecurityfrontier modelsWinfunc Research