Compositional Generalization on COGS
83.9Exact Match Accuracy
GRPO-Binary
Updated 27d ago
Evaluation Results
| Method | Links | |
|---|---|---|
2026.05 | 83.9 | |
2026.05 | 83.23 | |
2026.05 | 83.15 | |
2026.05 | 82.88 | |
2026.05 | 82.09 | |
2026.05 | 81.12 |
GRPO-Binary
| Method | Links | |
|---|---|---|
2026.05 | 83.9 | |
2026.05 | 83.23 | |
2026.05 | 83.15 | |
2026.05 | 82.88 | |
2026.05 | 82.09 | |
2026.05 | 81.12 |