Process-level Reward Modeling on PROCESSBENCH MATH
6.1Error Rate
SPARE-Llama3-8B
Updated 1mo ago
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
2025.06 | 6.1 | 91.6 | 11.4 | |
2025.06 | 16 | 89.2 | 27.1 | |
2025.06 | 18 | 82 | 29.5 | |
2025.06 | 21.4 | 80 | 33.8 | |
2025.06 | 43.8 | 62.2 | 53.6 | |
| 48 | 90.1 | 62.6 |
SPARE-Llama3-8B
| Method | Links | |||
|---|---|---|---|---|
2025.06 | 6.1 | 91.6 | 11.4 | |
2025.06 | 16 | 89.2 | 27.1 | |
2025.06 | 18 | 82 | 29.5 | |
2025.06 | 21.4 | 80 | 33.8 | |
2025.06 | 43.8 | 62.2 | 53.6 | |
| 48 | 90.1 | 62.6 |