Multi-step Reasoning on ScienceQA
92.72Accuracy
SceneAlign
Updated 4d ago
Evaluation Results
| Method | Links | |
|---|---|---|
2026.01 | 92.72 | |
2026.01 | 91.53 | |
2026.01 | 91.49 | |
2026.01 | 90.77 | |
2026.01 | 90.15 | |
2026.01 | 90.02 | |
2026.01 | 89.12 | |
2026.01 | 88.8 | |
2026.01 | 88.71 | |
2026.01 | 85.1 | |
2026.01 | 83.26 | |
2026.01 | 83.25 | |
2026.01 | 83.19 | |
2026.01 | 81.99 | |
2026.01 | 81.25 | |
2026.01 | 81.1 | |
2026.01 | 80.12 |