Self-correction on Qwen-72B math failure pool n=30
80Correction Rate
L_MEMORY (Source-Conditioned Role Relabeling)
Updated 1mo ago
Evaluation Results
| Method | Links | |
|---|---|---|
| 80 | ||
2026.06 | 46.7 | |
2026.06 | 30 | |
2026.06 | 10 | |
2026.06 | 6.7 |
L_MEMORY (Source-Conditioned Role Relabeling)
| Method | Links | |
|---|---|---|
| 80 | ||
2026.06 | 46.7 | |
2026.06 | 30 | |
2026.06 | 10 | |
2026.06 | 6.7 |