Persuasion on Anthropic Persuasion Dataset Persuadee: Phi-4 OOD
32.42Agreement ShiftLLaMa3.2 + ToMAP
Evaluation Results
| Method | Links | |
|---|---|---|
| LLaMa3.2 + ToMAPSize=3B2025.05 | 32.42 | |
| DS-R1Size=671B2025.05 | 30.61 | |
| LLaMa3.2 + RLSize=3B2025.05 | 30.27 | |
| Gemma-3Size=27B2025.05 | 29.91 | |
| GPT-4oSize=N/A2025.05 | 26.56 | |
| Qwen2.5Size=3B2025.05 | 23.24 | |
| Qwen2.5 + ToMAPSize=3B2025.05 | 22.79 | |
| Qwen2.5 + SFTSize=3B2025.05 | 22.35 | |
| LLaMa3.1Size=70B2025.05 | 20.5 | |
| LLaMa3.2 + SFTSize=3B2025.05 | 20.03 | |
| LLaMa3.2Size=3B2025.05 | 18.75 | |
| Qwen2.5 + RLSize=3B2025.05 | 16.84 |