List your company

Win rate evaluation on SLIMORCA (test)

88.12Win RateTOSS
48.475258.767669.0679.3524Mar 1, 2026

Evaluation results

MethodLinks
TOSSBackbone=Llama-3-8B-Instruct2026.0388.12
TOSSComparison model=Standard SFT, Backbone=Llama-3-8B-Instruct2026.0368.37
TOSSParadigm=Training-based defense, Base Model=Llama-3-8B-Instruct2026.0368.37
TOSSComparison model=SEAL, Backbone=Llama-3-8B-Instruct2026.0363.96
Circuit BreakersParadigm=Training-based defense, Base Model=Llama-3-8B-Instruct2026.0350.95
SafetyLockParadigm=Training-free defense, Base Model=Llama-3-8B-Instruct2026.0350.72
Standard SFTBackbone=Llama-3-8B-Instruct2026.0350
Standard SFTParadigm=/, Base Model=Llama-3-8B-Instruct2026.0350