Logical Reasoning on Countdown
52Accuracy
+ EOS Temp.
Updated 1mo ago
Evaluation Results
| Method | Links | |
|---|---|---|
2026.04 | 52 | |
2026.04 | 51.2 | |
2026.04 | 48.4 | |
2026.04 | 46.9 | |
2026.04 | 46.5 | |
2026.04 | 46.1 | |
2026.04 | 45.7 | |
2026.04 | 44.5 | |
2026.04 | 43.8 | |
2026.04 | 41.8 | |
2026.04 | 41.4 | |
2026.04 | 41.4 | |
2026.04 | 39.5 | |
2026.04 | 36.7 | |
2026.04 | 25.4 | |
2026.04 | 23.8 |