Persuasion on Cornell CMV Persuadee: Qwen2.5
26.89Agreement ShiftLLaMa3.2 + ToMAP
Evaluation Results
| Method | Links | |
|---|---|---|
| LLaMa3.2 + ToMAPBase Model=LLaMa3.2-3B, Size=3B, Training Method=ToMAP2025.05 | 26.89 | |
| Qwen2.5 + ToMAPBase Model=Qwen2.5-3B, Size=3B, Training Method=ToMAP2025.05 | 23.01 | |
| LLaMa3.2 + RLBase Model=LLaMa3.2-3B, Size=3B, Training Method=RL2025.05 | 21.34 | |
| LLaMa3.2 + ToMAPSize=3B2025.05 | 20.96 | |
| LLaMa3.2 + RLSize=3B2025.05 | 19.34 | |
| Qwen2.5 + ToMAPSize=3B2025.05 | 18.75 | |
| Qwen2.5 + RLBase Model=Qwen2.5-3B, Size=3B, Training Method=RL2025.05 | 17.42 | |
| DS-R1Size=671B, Training Method=Off-the-shelf2025.05 | 16.04 | |
| LLaMa3.2 + SFTBase Model=LLaMa3.2-3B, Size=3B, Training Method=SFT2025.05 | 14.9 | |
| LLaMa3.2Size=3B, Training Method=Off-the-shelf2025.05 | 14.4 | |
| Qwen2.5 + RLSize=3B2025.05 | 13.77 | |
| Gemma-3Size=27B, Training Method=Off-the-shelf2025.05 | 13.52 | |
| Qwen2.5 + SFTBase Model=Qwen2.5-3B, Size=3B, Training Method=SFT2025.05 | 13.41 | |
| GPT-4oSize=N/A, Training Method=Off-the-shelf2025.05 | 13.36 | |
| Qwen2.5Size=3B, Training Method=Off-the-shelf2025.05 | 12.75 | |
| LLaMa3.1Size=70B, Training Method=Off-the-shelf2025.05 | 12.58 | |
| Gemma-3Size=27B2025.05 | 12.21 | |
| DS-R1Size=671B2025.05 | 10.34 | |
| GPT-4oSize=N/A2025.05 | 9.51 | |
| LLaMa3.2 + SFTSize=3B2025.05 | 9.1 | |
| Qwen2.5 + SFTSize=3B2025.05 | 8.97 | |
| LLaMa3.2Size=3B2025.05 | 8.4 | |
| LLaMa3.1Size=70B2025.05 | 7.09 | |
| Qwen2.5Size=3B2025.05 | 6.9 |