Persuasion on Cornell CMV Persuadee: LLaMa3.1 OOD
20.84Agreement ShiftLLaMa3.2 + ToMAP
Evaluation Results
| Method | Links | |
|---|---|---|
| LLaMa3.2 + ToMAPSize=3B2025.05 | 20.84 | |
| LLaMa3.2 + RLSize=3B2025.05 | 19.16 | |
| LLaMa3.2 + RLSize=3B2025.05 | 17.62 | |
| LLaMa3.2 + ToMAPSize=3B2025.05 | 17.57 | |
| Qwen2.5 + RLSize=3B2025.05 | 12.89 | |
| Qwen2.5 + RLSize=3B2025.05 | 12.15 | |
| Gemma-3Size=27B2025.05 | 11.35 | |
| Qwen2.5 + ToMAPSize=3B2025.05 | 10.89 | |
| Qwen2.5 + ToMAPSize=3B2025.05 | 10.77 | |
| Gemma-3Size=27B2025.05 | 9.69 | |
| LLaMa3.2 + SFTSize=3B2025.05 | 8.89 | |
| DS-R1Size=671B2025.05 | 8.51 | |
| DS-R1Size=671B2025.05 | 7.69 | |
| LLaMa3.2Size=3B2025.05 | 7.56 | |
| LLaMa3.2Size=3B2025.05 | 7.22 | |
| Qwen2.5 + SFTSize=3B2025.05 | 6.51 | |
| LLaMa3.2 + SFTSize=3B2025.05 | 6.18 | |
| Qwen2.5Size=3B2025.05 | 4.97 | |
| Qwen2.5 + SFTSize=3B2025.05 | 4.34 | |
| Qwen2.5Size=3B2025.05 | 4.09 | |
| GPT-4oSize=N/A2025.05 | 3.35 | |
| GPT-4oSize=N/A2025.05 | 2.5 | |
| LLaMa3.1Size=70B2025.05 | 2.23 | |
| LLaMa3.1Size=70B2025.05 | 2.22 |