List your company

Adversarial Attack Robustness on BeaverTails

0Attack Success RateBase+ART
-3.8422.084873.92May 26, 2026

Evaluation results

MethodLinks
Base+ARTModel=Gemma3-1B-IT, Attack Scenario=Baseline2026.050
SEAM+ARTModel=Gemma3-1B-IT, Attack Scenario=Baseline2026.050
TAR+ARTModel=Gemma3-1B-IT, Attack Scenario=Baseline2026.050
TAR+ARTModel=Gemma3-1B-IT, Attack Scenario=Prefill2026.051
SEAM+ARTModel=Gemma3-1B-IT, Attack Scenario=Prefill2026.052
TARModel=Gemma3-1B-IT, Attack Scenario=Prefill2026.054
Base+ARTModel=Gemma3-1B-IT, Attack Scenario=Prefill2026.055
TARModel=Gemma3-1B-IT, Attack Scenario=Baseline2026.055
BaseModel=Gemma3-1B-IT, Attack Scenario=Baseline2026.056
SEAMModel=Gemma3-1B-IT, Attack Scenario=Baseline2026.057
SEAMModel=Gemma3-1B-IT, Attack Scenario=Prefill2026.0516
BaseModel=Gemma3-1B-IT, Attack Scenario=Prefill2026.0520
Base+ARTModel=Gemma3-1B-IT, Attack Scenario=Abliterate2026.0549
TAR+ARTModel=Gemma3-1B-IT, Attack Scenario=Abliterate2026.0555
TAR+ARTModel=Gemma3-1B-IT, Attack Scenario=Abliterate + Prefill2026.0556
SEAM+ARTModel=Gemma3-1B-IT, Attack Scenario=Abliterate2026.0564
TARModel=Gemma3-1B-IT, Attack Scenario=Abliterate + Prefill2026.0566
Base+ARTModel=Gemma3-1B-IT, Attack Scenario=Abliterate + Prefill2026.0567
SEAM+ARTModel=Gemma3-1B-IT, Attack Scenario=Abliterate + Prefill2026.0568
SEAMModel=Gemma3-1B-IT, Attack Scenario=Abliterate + Prefill2026.0576
TARModel=Gemma3-1B-IT, Attack Scenario=Abliterate2026.0576
SEAMModel=Gemma3-1B-IT, Attack Scenario=Abliterate2026.0580
BaseModel=Gemma3-1B-IT, Attack Scenario=Abliterate2026.0593
BaseModel=Gemma3-1B-IT, Attack Scenario=Abliterate + Prefill2026.0596