Capture The Flag on NYU-CTF (test)
2Delta vs KryptoPilotKryptoPilot
Evaluation results
| Method | Links | ||
|---|---|---|---|
| KryptoPilotVariant=HT-enhanced (HeavyThink mode)2026.01 | 2 | — | |
| KryptoPilotVariant=w/o DR (Deep Research)2026.01 | -3 | — | |
| CTFAgentVariant=baseline2026.01 | -8 | — |