Preference Alignment on AlpacaEval
52Win RateAdaBoN
Evaluation Results
| Method | Links | |
|---|---|---|
| AdaBoNLM=Mistral, RM=FsfairX, K=5, B=120, d=0.75B2025.05 | 52 | |
| AdaBoNLM=Mistral, RM=Armo, K=5, B=120, d=0.75B2025.05 | 52 | |
| AdaBoNLM=Llama, RM=Mistral, K=5, B=120, d=0.75B2025.05 | 52 | |
| AdaBoNLM=Llama, RM=FsfairX, K=5, B=120, d=0.75B2025.05 | 52 | |
| AdaBoNLM=Llama, RM=Armo, K=5, B=120, d=0.75B2025.05 | 52 | |
| AdaBoNLM=Mistral, RM=Mistral, K=5, B=120, d=0.75B2025.05 | 51 | |
| AdaBoNLM=Qwen, RM=Mistral, K=5, B=120, d=0.75B2025.05 | 51 | |
| AdaBoNLM=Qwen, RM=FsfairX, K=5, B=120, d=0.75B2025.05 | 51 | |
| AdaBoNLM=Qwen, RM=Armo, K=5, B=120, d=0.75B2025.05 | 51 | |
| AdaBoNLM=Gemma, RM=Mistral, K=5, B=120, d=0.75B2025.05 | 51 | |
| AdaBoNLM=Gemma, RM=FsfairX, K=5, B=120, d=0.75B2025.05 | 51 | |
| AdaBoNLM=Gemma, RM=Armo, K=5, B=120, d=0.75B2025.05 | 51 |