Mathematical Reasoning on CHAMP
68.2AccuracyMDToC
Evaluation Results
| Method | Links | |
|---|---|---|
| MDToCbackbone=GPT-4o2025.12 | 68.2 | |
| 1/3 slnbackbone=GPT-4o2025.12 | 63.5 | |
| ToTbackbone=GPT-4o2025.12 | 61.3 | |
| GoTbackbone=GPT-4o2025.12 | 60.9 | |
| CoT + CHbackbone=GPT-4o2025.12 | 60 | |
| MDToCbackbone=GPT-4-Turbo2025.12 | 58.1 | |
| 5-shotbackbone=GPT-4o2025.12 | 56.5 | |
| 0-shotbackbone=GPT-4o2025.12 | 55.3 | |
| CoTbackbone=GPT-4o2025.12 | 55.2 | |
| 1/3 slnbackbone=GPT-4-Turbo2025.12 | 53.7 | |
| GoTbackbone=GPT-4-Turbo2025.12 | 53.1 | |
| CoT + CHbackbone=GPT-4-Turbo2025.12 | 53 | |
| ToTbackbone=GPT-4-Turbo2025.12 | 52.7 | |
| MDToCbackbone=GPT-4o-mini2025.12 | 48.3 | |
| 1/3 slnbackbone=GPT-4o-mini2025.12 | 43.4 | |
| 5-shotbackbone=GPT-4-Turbo2025.12 | 43.1 | |
| CoT + CHbackbone=GPT-4o-mini2025.12 | 42.3 | |
| 0-shotbackbone=GPT-4-Turbo2025.12 | 41.9 | |
| GoTbackbone=GPT-4o-mini2025.12 | 39.6 | |
| MDToCbackbone=GPT-3.5-Turbo2025.12 | 39.5 | |
| 5-shotbackbone=GPT-4o-mini2025.12 | 38.7 | |
| CoTbackbone=GPT-4-Turbo2025.12 | 37.8 | |
| ToTbackbone=GPT-4o-mini2025.12 | 37.8 | |
| CoTbackbone=GPT-4o-mini2025.12 | 36.5 | |
| 0-shotbackbone=GPT-4o-mini2025.12 | 36.2 | |
| 5-shotbackbone=GPT-3.5-Turbo2025.12 | 34.8 | |
| CoT + CHbackbone=GPT-3.5-Turbo2025.12 | 34.4 | |
| 1/3 slnbackbone=GPT-3.5-Turbo2025.12 | 33.7 | |
| ToTbackbone=GPT-3.5-Turbo2025.12 | 31.7 | |
| GoTbackbone=GPT-3.5-Turbo2025.12 | 30.5 | |
| CoTbackbone=GPT-3.5-Turbo2025.12 | 29.6 | |
| 0-shotbackbone=GPT-3.5-Turbo2025.12 | 28.5 |