General Reasoning on Average (MATH500, OlympiadBench, Minerva, MMLU, GSM8K)
70Average AccuracyS-BoN (b)
Evaluation results
| Method | Links | |
|---|---|---|
| S-BoN (b)n=2562025.06 | 70 | |
| S-BoN (b)n=642025.06 | 69.6 | |
| S-BoN (b)n=162025.06 | 69.4 | |
| S-BoN (b)n=42025.06 | 68.9 | |
| S-BoN (b)n=12025.06 | 67.5 | |
| GSIn=2562025.06 | 67.2 | |
| GSIn=642025.06 | 67 | |
| GSIn=162025.06 | 66.9 | |
| GSIn=42025.06 | 65.2 | |
| GSIn=12025.06 | 61.5 | |
| RSDn=642025.06 | 61.5 | |
| RSDn=2562025.06 | 61.1 | |
| RSDn=162025.06 | 60.9 | |
| RSDn=42025.06 | 60.8 | |
| S-BoN (s)n=2562025.06 | 60.6 | |
| S-BoN (s)n=642025.06 | 59.7 | |
| S-BoN (s)n=162025.06 | 58.5 | |
| RSDn=12025.06 | 57.4 | |
| S-BoN (s)n=42025.06 | 56.6 | |
| S-BoN (s)n=12025.06 | 51.5 |