| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Goal Conditioned Visual Navigation | SCAND | ATE1.065 | 18 | |
| Geometric Drift Evaluation | SCAND | Endpoint Distance (ED)3.58 | 15 | |
| Perceptual Drift | SCAND | LPIPS0.297 | 15 | |
| Visual Fidelity | SCAND | FID80.97 | 15 | |
| Trajectory Prediction | SCAND | ATE1.14 | 12 | |
| Long-horizon prediction | SCAND | LPIPS0.396 | 10 | |
| Trajectory Prediction | SCAND T=8 | L2 Error (m)0.923 | 10 | |
| Trajectory Prediction | SCAND T=5 | L2 Error (m)0.674 | 10 | |
| Action Prediction | SCAND (test) | Action MSE0.47 | 8 | |
| Goal-Conditioned Visual Navigation | SCAND (evaluation) | Success Rate (SR)68 | 8 | |
| Trajectory Prediction | SCAND T=10 | L2 Error (m)1.089 | 8 | |
| Social Navigation | SCAND Out-of-Distribution (test) | ADE0.578 | 7 | |
| Social Navigation | SCAND 4,902 frames (evaluation) | NC2,580 | 6 | |
| Video Synthesis | SCAND | PSNR16.11 | 6 | |
| Visual Navigation | SCAND | MSE0.48 | 5 | |
| Planning | SCAND | ATE2.3 | 4 | |
| Navigation Planning | Scand (val) | ATE1.038 | 3 | |
| Action-Conditioned Consistency | SCAND 16s horizon | LPIPS0.495 | 3 | |
| Action-Conditioned Consistency | SCAND 8s horizon | LPIPS0.459 | 3 | |
| Action-Conditioned Consistency | SCAND 4s horizon | LPIPS0.421 | 3 | |
| Action-Conditioned Consistency | SCAND 2s horizon | LPIPS0.395 | 3 | |
| Action-Conditioned Consistency | SCAND 1s horizon | LPIPS0.368 | 3 | |
| Inference Efficiency | SCAND | Average Rollout Time (s)2.3 | 3 | |
| Goal-Conditioned Visual Navigation | SCAND 16s horizon | ATE18.38 | 2 | |
| Goal-Conditioned Visual Navigation | SCAND 8s horizon | ATE11.33 | 2 |