Safety evaluation on Safety Eval
93.8Safety ScoreEvoPref-Best
Evaluation Results
| Method | Links | |
|---|---|---|
| EvoPref-BestOptimization Paradigm=Multi-Objective Evolutionary, Base Model=Mistral-7B-Instruct-v0.2, LoRA Rank=16, LoRA Alpha=322026.05 | 93.8 | |
| EvoPrefOptimization Paradigm=Multi-Objective Evolutionary, Base Model=Mistral-7B-Instruct-v0.2, LoRA Rank=16, LoRA Alpha=322026.05 | 91.2 | |
| SMS-EMOAOptimization Paradigm=Multi-Objective Evolutionary, Base Model=Mistral-7B-Instruct-v0.2, LoRA Rank=16, LoRA Alpha=322026.05 | 90.7 | |
| MOEA/DOptimization Paradigm=Multi-Objective Evolutionary, Base Model=Mistral-7B-Instruct-v0.2, LoRA Rank=16, LoRA Alpha=322026.05 | 90.1 | |
| KTOOptimization Paradigm=Gradient-Based, Base Model=Mistral-7B-Instruct-v0.2, LoRA Rank=16, LoRA Alpha=322026.05 | 90 | |
| ORPOOptimization Paradigm=Gradient-Based, Base Model=Mistral-7B-Instruct-v0.2, LoRA Rank=16, LoRA Alpha=322026.05 | 89.7 | |
| DPOOptimization Paradigm=Gradient-Based, Base Model=Mistral-7B-Instruct-v0.2, LoRA Rank=16, LoRA Alpha=322026.05 | 89.2 | |
| CMA-ESOptimization Paradigm=Single-Objective Evolutionary, Base Model=Mistral-7B-Instruct-v0.2, LoRA Rank=16, LoRA Alpha=322026.05 | 88.8 | |
| IPOOptimization Paradigm=Gradient-Based, Base Model=Mistral-7B-Instruct-v0.2, LoRA Rank=16, LoRA Alpha=322026.05 | 88.5 |