CVE-Bench
Named by 3 kept items in this workspace.
2 items name it under tools and 1 under classification.key entities.
Search every workspace for CVE-Bench. It matches the name as written, so it can find other things that share it.
The items that name it 3
Benchmark tasking models with identifying and exploiting real-world web-application vulnerabilities in a sandbox, run zero-day configuration without source code access (34 of 40 challenges run)
2025 real-CVE evaluation benchmark
All technologies in this workspace.