CyberBinGym

Binary patch-gap benchmark with separate vulnerability-recovery and end-to-end exploitation checks.

Records citing CyberBinGym

Nov 7, 2026
CyberBinGym separates binary vulnerability recovery from working exploitation
CapabilityBenchmarkWijayadi et al. / arXiv

The CyberBinGym preprint evaluates agents given stripped vulnerable and patched binaries for 100 tasks from 53 projects. On the non-hardened baseline, Codex with GPT-5.6 recovers vulnerabilities on 88% of tasks and completes exploitation on 20%. The known-vulnerability patch-gap tests use trusted-access models with safeguards disabled; hardened configurations are evaluated separately.