| Dataset Name | SOTA Method | Metric | Trend | ||
|---|---|---|---|---|---|
| Maze | DiffThinker | Score (L=8)100 | 12 | 4mo ago | |
| VSP-Super | DiffThinker++ | Success Rate (Length 16)99 | 12 | 4mo ago | |
| Visual Spatial Planning (VSP) FrozenLake | VSP FrozenLake Success Rate (Level 3)100 | 12 | 4mo ago | ||
| PlanCraft | Multi-agent + sequential | Performance Change vs. SAS39.1 | 1 | 18d ago |