| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Cybersecurity vulnerability remediation | CVE-Bench (one-day) | Pass@129 | 5 | |
| Automated Vulnerability Exploitation | CVE-Bench zero-day vulnerabilities (test) | Success@125 | 5 | |
| Cybersecurity vulnerability remediation | CVE-Bench | Pass@125 | 2 | |
| Cybersecurity vulnerability remediation | CVE-Bench 1.0 | Pass@1 Rate90 | 1 | |
| Cybersecurity vulnerability remediation | CVE-Bench (zero-day) | Pass@18 | 1 |