| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Synthetic Long-context Reasoning | GraphWalks | Accuracy68.91 | 12 | |
| Long Context Performance | Graphwalks Bfs (<128K) | Score98 | 9 | |
| Graph traversal | GraphWalks | Parents Score89.87 | 8 | |
| Long-context graph reasoning | GraphWalks | F1 Score82.3 | 6 | |
| Long Context Performance | Graphwalks Parents | Score99.7 | 5 | |
| Long Context Performance | Graphwalks Parents (<128K) | Score100 | 4 |