Mathematical Reasoning on AQuA (Verifier Settings)
74Accuracy (Without Verifier)ToT
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| ToTModel=Llama-3.1-70B2025.03 | 74 | 74 | 74 | |
| CoTModel=Llama-3.1-70B2025.03 | 72 | 73 | 73 | |
| L2MModel=Llama-3.1-70B2025.03 | 72 | 72 | 73 | |
| MCTSModel=Llama-3.1-70B2025.03 | 72 | 73 | 72 | |
| CoTModel=Llama-3.1-8B2025.03 | 60 | 63 | 60 | |
| L2MModel=Llama-3.1-8B2025.03 | 58 | 62 | 58 | |
| ToTModel=Llama-3.1-8B2025.03 | 50 | 53 | 50 | |
| MCTSModel=Llama-3.1-8B2025.03 | 50 | 51 | 50 | |
| CoTModel=Llama-3.2-3B2025.03 | 46 | 51 | 46 | |
| ToTModel=Llama-3.2-1B2025.03 | 35 | 38 | 35 | |
| MCTSModel=Llama-3.2-3B2025.03 | 35 | 36 | 35 | |
| ToTModel=Llama-3.2-3B2025.03 | 33 | 35 | 33 | |
| MCTSModel=Llama-3.2-1B2025.03 | 29 | 32 | 31 | |
| L2MModel=Llama-3.2-3B2025.03 | 29 | 31 | 31 | |
| CoTModel=Llama-3.2-1B2025.03 | 26 | 28 | 26 | |
| L2MModel=Llama-3.2-1B2025.03 | 22 | 24 | 29 |