List your company

Red-teaming Safety Evaluation on StrongReject

4ASRTAP
0.185625.932851.6877.4272May 30, 2025Jul 23, 2025Sep 15, 2025Nov 9, 2025Jan 2, 2026Feb 25, 2026Apr 21, 2026

Evaluation results

MethodLinks
TAPTarget Model=Llama2-7b2026.044——
PAIRTarget Model=Llama2-7b2026.045——
PAP-top5Target Model=Gemma-7b2026.046——
Meta-Llama-3.1-8B (ATTAQ-aligned)base_model=Meta-Llama-3.1-8B, alignment_dataset=ATTAQ2025.057111.79
Meta-Llama-3.1-8B (WILDBREAK-aligned)base_model=Meta-Llama-3.1-8B, alignment_dataset=WILDBREAK2025.057182.14
PAIRTarget Model=Gemma-7b2026.048——
PAP-top5Target Model=Llama3-8b2026.048——
Rainbow TeamingTarget Model=Llama2-7b2026.048——
Rainbow TeamingTarget Model=Gemma-7b2026.048——
Meta-Llama-3.1-8B (TRIDENT-EDGE-aligned)base_model=Meta-Llama-3.1-8B, alignment_dataset=TRIDENT-EDGE2025.05981.86
Rainbow TeamingTarget Model=Llama3-8b2026.049——
Meta-Llama-3.1-8B (AART-aligned)base_model=Meta-Llama-3.1-8B, alignment_dataset=AART2025.0510131.94
GCG-TTarget Model=Llama3-8b2026.0410——
GCG-TTarget Model=Gemma-7b2026.0410——
PAP-top5Target Model=Llama2-7b2026.0410——
GCG-TTarget Model=Llama2-7b2026.0412——
PAIRTarget Model=Llama3-8b2026.0412——
TAPTarget Model=Llama3-8b2026.0413——
TAPTarget Model=Gemma-7b2026.0416——
GCGTarget Model=Llama3.1-8B2026.0118.27——
Meta-Llama-3.1-8B (SAFE_RLHF-aligned)base_model=Meta-Llama-3.1-8B, alignment_dataset=SAFE_RLHF2025.0519212.41
Meta-Llama-3.1-8B (WILDCHAT-aligned)base_model=Meta-Llama-3.1-8B, alignment_dataset=WILDCHAT2025.0520232.48
AutoDANTarget Model=Llama2-7B2026.0120.45——
GCGTarget Model=Llama2-7B2026.0120.77——
Meta-Llama-3.1-8B (HH_RLHF-aligned)base_model=Meta-Llama-3.1-8B, alignment_dataset=HH_RLHF2025.0523332.47
Meta-Llama-3.1-8B (Unaligned)base_model=Meta-Llama-3.1-8B2025.0525692.66
GCGTarget Model=Qwen2.5-7B2026.0134.82——
STAR-TeamingAttacker Model=Llama3-8b, Target Model=Llama3-8b2026.0443——
STAR-TeamingAttacker Model=Gemma-7b, Target Model=Llama3-8b2026.0445——
STAR-TeamingAttacker Model=Llama3-8b, Target Model=Llama2-7b2026.0450——
STAR-TeamingAttacker Model=Gemma-7b, Target Model=Gemma-7b2026.0455——
DSNTarget Model=Qwen2.5-7B2026.0156.87——
STAR-TeamingAttacker Model=Gemma-7b, Target Model=Llama2-7b2026.0457——
AutoDANTarget Model=Llama3.1-8B2026.0157.51——
DSNTarget Model=Llama2-7B2026.0159.11——
DSNTarget Model=Llama3.1-8B2026.0160.38——
STAR-TeamingAttacker Model=Llama3-8b, Target Model=Gemma-7b2026.0461——
AdvPrefixTarget Model=Llama2-7B2026.0165.18——
ORTHOTarget Model=Llama2-7B2026.0166.77——
DSNTarget Model=Mistral-7B2026.0171.57——
AutoDANTarget Model=Mistral-7B2026.0173.48——
ORTHOTarget Model=Mistral-7B2026.0174.44——
AutoDANTarget Model=Qwen2.5-7B2026.0175.08——
AdvPrefixTarget Model=Llama3.1-8B2026.0175.4——
AdvPrefixTarget Model=Qwen2.5-7B2026.0177.96——
GCGTarget Model=Mistral-7B2026.0182.43——
ORTHOTarget Model=Llama3.1-8B2026.0187.22——
ORTHOTarget Model=Qwen2.5-7B2026.0188.5——
AdvPrefixTarget Model=Mistral-7B2026.0190.1——
GOSVTarget Model=Qwen2.5-7B2026.0190.42——
GOSVTarget Model=Llama2-7B2026.0191.05——
GOSVTarget Model=Llama3.1-8B2026.0192.01——
GOSVTarget Model=Mistral-7B2026.0199.36——