Mathematical Reasoning on MR-GSM8K
92AccuracyToT
Evaluation Results
| Method | Links | |
|---|---|---|
| ToTBase Model=Claude 3.5 Sonnet2024.10 | 92 | |
| DIDBase Model=Claude 3.5 Sonnet2024.10 | 92 | |
| CoTBase Model=Claude 3.5 Sonnet2024.10 | 91.3 | |
| ToTBase Model=GPT-4o2024.10 | 89.1 | |
| DIDBase Model=GPT-4o2024.10 | 87.7 | |
| CoTBase Model=GPT-4o2024.10 | 85 | |
| ToTBase Model=GPT-3.5 Turbo2024.10 | 74 | |
| DIDBase Model=GPT-3.5 Turbo2024.10 | 73.3 | |
| CoTBase Model=GPT-3.5 Turbo2024.10 | 68.1 |