Causal Reasoning on CRASS OOD (held-out)
95Performance P0
Qwen2.5-32B
Updated 1mo ago
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
2026.06 | 95 | 72.5 | 22.5 | -18 | |
2026.06 | 95 | 72.5 | 22.5 | -18 | |
2026.06 | 91.8 | 72.2 | 19.7 | -27 | |
2026.06 | 91.8 | 72.2 | 19.7 | -27 | |
2026.06 | 91 | 63.5 | 27.5 | - | |
2026.06 | 91 | 63.5 | 27.5 | - | |
2026.06 | 89.5 | 62.5 | 27 | - | |
2026.06 | 89.5 | 62.5 | 27 | - | |
2026.06 | 87.2 | 55.8 | 31.3 | -23 | |
2026.06 | 87.2 | 55.8 | 31.3 | -23 | |
2026.06 | 87 | 62.2 | 24.8 | -27 | |
2026.06 | 87 | 62.2 | 24.8 | -27 | |
2026.06 | 85 | 44.5 | 40.5 | - | |
2026.06 | 85 | 44.5 | 40.5 | - | |
2026.06 | 84.5 | 54.5 | 30 | - | |
2026.06 | 84.5 | 50.5 | 34 | - | |
2026.06 | 84.5 | 54.5 | 30 | - | |
2026.06 | 84.5 | 50.5 | 34 | - | |
2026.06 | 84 | 55 | 29 | -3 | |
2026.06 | 84 | 55 | 29 | -3 | |
2026.06 | 79.5 | 41.5 | 38 | - | |
2026.06 | 79.5 | 41.5 | 38 | - | |
2026.06 | 79.5 | 41.5 | 38 | - | |
2026.06 | 73.5 | 36.5 | 37 | -3 | |
2026.06 | 73.5 | 36.5 | 37 | -3 | |
2026.06 | 71 | 50 | 21 | -45 | |
2026.06 | 71 | 50 | 21 | -45 |