List your company

Code Safety Evaluation on BigCodeBench 1.0 (test)

99.9Safety ScoreSelf-Rec (No Elicitation)
2.97228.13653.378.464Feb 24, 2026

Evaluation results

MethodLinks
Self-Rec (No Elicitation)Signal Jamming=None, Monitoring=UM+TM2026.0299.9
Self-Rec (No Elicitation)Signal Jamming=Minifier, Monitoring=UM+TM2026.0299.9
Self-Rec (No Elicitation)Signal Jamming=Summariser, Monitoring=UM+TM2026.0299.9
Vague AcausalSignal Jamming=None, Monitoring=UM+TM2026.0299.9
Vague AcausalSignal Jamming=Summariser, Monitoring=UM+TM2026.0299.9
Causal (For/While)Signal Jamming=Summariser, Monitoring=UM+TM2026.0299.9
Self-Rec (SFT)Signal Jamming=Minifier, Monitoring=UM+TM2026.0299.6
Causal (For/While)Signal Jamming=None, Monitoring=UM+TM2026.0299.6
Causal (For/While)Signal Jamming=Minifier, Monitoring=UM+TM2026.0299.6
Vague AcausalSignal Jamming=Minifier, Monitoring=UM+TM2026.0299.5
Self-Rec (SFT)Signal Jamming=None, Monitoring=UM+TM2026.0299.3
Self-Rec (SFT)Signal Jamming=Summariser, Monitoring=UM+TM2026.0299.2
Self-Rec (No Elicitation)Signal Jamming=None, Monitoring=UM-only2026.0297.7
Self-Rec (No Elicitation)Signal Jamming=Minifier, Monitoring=UM-only2026.0297.6
Self-Rec (No Elicitation)Signal Jamming=Summariser, Monitoring=UM-only2026.0296.2
Vague AcausalSignal Jamming=Minifier, Monitoring=UM-only2026.0295
Causal (For/While)Signal Jamming=Summariser, Monitoring=UM-only2026.0291.7
Vague AcausalSignal Jamming=Summariser, Monitoring=UM-only2026.0287.8
Vague AcausalSignal Jamming=None, Monitoring=UM-only2026.0273.2
Self-Rec (SFT)Signal Jamming=Summariser, Monitoring=UM-only2026.0273
Self-Rec (SFT)Signal Jamming=Minifier, Monitoring=UM-only2026.0268.6
Causal (For/While)Signal Jamming=Minifier, Monitoring=UM-only2026.0213.3
Self-Rec (SFT)Signal Jamming=None, Monitoring=UM-only2026.0211.1
Causal (For/While)Signal Jamming=None, Monitoring=UM-only2026.026.7