List your company

Safety Evaluation on FigStep (test)

0ASRLLaVA-1.5-7B + Mixed-SFT
-3.4419.784366.22Mar 14, 2025

Evaluation results

MethodLinks
LLaVA-1.5-7B + Mixed-SFTAttack Status=Before, Tuning Strategy=Full-parameter2025.030
LLaVA-1.5-7B + Posthoc-SFTAttack Status=Before, Tuning Strategy=Full-parameter2025.030
LLaVA-1.5-7B-LoRA + Mixed-SFTAttack Status=Before, Tuning Strategy=LoRA2025.030
LLaVA-1.5-7B-LoRA + Posthoc-SFTAttack Status=Before, Tuning Strategy=LoRA2025.030
LLaVA-1.5-7B-LoRA + RMU-UnlearningAttack Status=Before, Tuning Strategy=LoRA2025.032
LLaVA-1.5-7B + RMU-UnlearningAttack Status=Before, Tuning Strategy=Full-parameter2025.034
LLaVA-1.5-7B-LoRA + NPO-UnlearningAttack Status=Before, Tuning Strategy=LoRA2025.034
LLaVA-1.5-7B + NPO-UnlearningAttack Status=Before, Tuning Strategy=Full-parameter2025.036
LLaVA-1.5-7B + RMU-UnlearningAttack Status=After, Tuning Strategy=Full-parameter2025.0310
LLaVA-1.5-7B + NPO-UnlearningAttack Status=After, Tuning Strategy=Full-parameter2025.0312
LLaVA-1.5-7B-LoRA + RMU-UnlearningAttack Status=After, Tuning Strategy=LoRA2025.0312
LLaVA-1.5-7B-LoRA + NPO-UnlearningAttack Status=After, Tuning Strategy=LoRA2025.0316
LLaVA-1.5-7B + Mixed-SFTAttack Status=After, Tuning Strategy=Full-parameter2025.0320
LLaVA-1.5-7B + Posthoc-SFTAttack Status=After, Tuning Strategy=Full-parameter2025.0328
LLaVA-1.5-7B-LoRA + Posthoc-SFTAttack Status=After, Tuning Strategy=LoRA2025.0336
LLaVA-1.5-7B-LoRA + Mixed-SFTAttack Status=After, Tuning Strategy=LoRA2025.0340
LLaVA-1.5-7BAttack Status=Before, Tuning Strategy=Full-parameter2025.0362
LLaVA-1.5-7B + Unsafe-FilterAttack Status=Before, Tuning Strategy=Full-parameter2025.0362
LLaVA-1.5-7B-LoRAAttack Status=Before, Tuning Strategy=LoRA2025.0372
LLaVA-1.5-7B-LoRA + Unsafe-FilterAttack Status=Before, Tuning Strategy=LoRA2025.0374
LLaVA-1.5-7B + Unsafe-FilterAttack Status=After, Tuning Strategy=Full-parameter2025.0384
LLaVA-1.5-7B-LoRAAttack Status=After, Tuning Strategy=LoRA2025.0384
LLaVA-1.5-7B-LoRA + Unsafe-FilterAttack Status=After, Tuning Strategy=LoRA2025.0384
LLaVA-1.5-7BAttack Status=After, Tuning Strategy=Full-parameter2025.0386