List your company

Safety Guardrailing on ULSPB 350 interaction runs

0.02HS RateStateGuard (Base-Ensemble)
-0.2361.4923.224.948May 7, 2026

Evaluation results

MethodLinks
StateGuard (Base-Ensemble)Backbone Model=GPT-5.4, Defense Setting=Base, Auditor Mode=Ensemble2026.050.025810
StateGuard (Targeted-Ensemble)Backbone Model=Grok 4.20, Defense Setting=Targeted, Auditor Mode=Ensemble2026.050.03614
StateGuard (Targeted-Ensemble)Backbone Model=GPT-5.4, Defense Setting=Targeted, Auditor Mode=Ensemble2026.050.05594
StateGuard (Targeted-Ensemble)Backbone Model=Kimi K2.5, Defense Setting=Targeted, Auditor Mode=Ensemble2026.050.06595
StateGuard (Base-Ensemble)Backbone Model=Grok 4.20, Defense Setting=Base, Auditor Mode=Ensemble2026.050.16113
StateGuard (Base-Single)Backbone Model=GPT-5.4, Defense Setting=Base, Auditor Mode=Single-Auditor2026.050.115714
StateGuard (Targeted-Single)Backbone Model=Grok 4.20, Defense Setting=Targeted, Auditor Mode=Single-Auditor2026.050.11588
StateGuard (Base-Ensemble)Backbone Model=Kimi K2.5, Defense Setting=Base, Auditor Mode=Ensemble2026.050.125612
StateGuard (Targeted-Ensemble)Backbone Model=MiniMax M2.7, Defense Setting=Targeted, Auditor Mode=Ensemble2026.050.14495
StateGuard (Targeted-Single)Backbone Model=Kimi K2.5, Defense Setting=Targeted, Auditor Mode=Single-Auditor2026.050.23578
StateGuard (Targeted-Single)Backbone Model=GPT-5.4, Defense Setting=Targeted, Auditor Mode=Single-Auditor2026.050.23597
StateGuard (Base-Single)Backbone Model=Grok 4.20, Defense Setting=Base, Auditor Mode=Single-Auditor2026.050.276017
StateGuard (Base-Single)Backbone Model=Kimi K2.5, Defense Setting=Base, Auditor Mode=Single-Auditor2026.050.365616
StateGuard (Targeted-Single)Backbone Model=MiniMax M2.7, Defense Setting=Targeted, Auditor Mode=Single-Auditor2026.050.37487
StateGuard (Base-Ensemble)Backbone Model=MiniMax M2.7, Defense Setting=Base, Auditor Mode=Ensemble2026.050.475211
StateGuard (Base-Single)Backbone Model=MiniMax M2.7, Defense Setting=Base, Auditor Mode=Single-Auditor2026.050.955415
ASB-PPLBackbone Model=Grok 4.202026.054.04693
ASB-PPLBackbone Model=Kimi K2.52026.054.25694
No DefenseBackbone Model=Grok 4.202026.054.35——
No DefenseBackbone Model=Kimi K2.52026.054.42——
ASB-PPLBackbone Model=GPT-5.42026.055.17990
No DefenseBackbone Model=GPT-5.42026.055.53——
ASB-PPLBackbone Model=MiniMax M2.72026.056.33196
No DefenseBackbone Model=MiniMax M2.72026.056.42——