List your company

Computer Use Control-Flow Hijacking on CFH-Hard Computer Use

67Gen. RateUndefended
-2.6815.4133.551.59Oct 20, 2025

Evaluation results

MethodLinks
Undefended2025.106783898978
ACF2025.106783898978
LP2025.106783898978
LlamaFirewallBackbone=llama2025.106783898978
LlamaFirewallBackbone=4o-mini2025.106778787872
LlamaFirewallBackbone=4o2025.105672788372
LlamaFirewallBackbone=o4-mini2025.10639721150
CONTROLVALVE2025.1000000