List your company

Safety Evaluation on HVD Explicit

0HC RateClaude-3.7-Sonnet
-10.336459.4343129.205198.9757Jun 4, 2026

Evaluation results

MethodLinks
Claude-3.7-SonnetAttack=None2026.06053.56
GPT-4oAttack=None2026.060.8871.92
DeepSeek-V3Attack=None2026.063.5482.85
GHOSTWRITERTarget Model=Claude-3.7-Sonnet, Attacker LM=GPT-4o-mini2026.0662.8392.73
GHOSTWRITERTarget Model=GPT-4o, Attacker LM=GPT-4o-mini2026.0672.5784.91
GHOSTWRITERTarget Model=DeepSeek-V3, Attacker LM=GPT-4o-mini2026.0683.1993
GHOSTWRITERTarget Model=Claude-3.7-Sonnet, Attacker LM=Qwen-2.5-7B2026.06164.690.41
GHOSTWRITERTarget Model=GPT-4o, Attacker LM=Qwen-2.5-7B2026.06202.6587.12
GHOSTWRITERTarget Model=DeepSeek-V3, Attacker LM=Qwen-2.5-7B2026.06258.4193.42