Goal-conditioned visual planning on COIN T=4 71
27.79SRGeoWorld
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| GeoWorldBackbone=ViT-g3842026.02 | 27.79 | 55.97 | 88.61 | |
| GeoWorldBackbone=ViT-g2026.02 | 27.46 | 54.55 | 88.2 | |
| GeoWorldBackbone=ViT-H2026.02 | 26.24 | 53.66 | 87.17 | |
| GeoWorldBackbone=ViT-L2026.02 | 24.96 | 52.92 | 85.26 | |
| V-JEPA 2Backbone=ViT-g3842026.02 | 23.43 | 55.58 | 86.57 | |
| V-JEPA 2Backbone=ViT-g2026.02 | 23.04 | 54.47 | 86.13 | |
| MTID2026.02 | 22.74 | 49.9 | 61.25 | |
| V-JEPA 2Backbone=ViT-H2026.02 | 22.6 | 53.19 | 85.74 | |
| SCHEMA2026.02 | 22.02 | 45.33 | 83.47 | |
| V-JEPA 2Backbone=ViT-L2026.02 | 20.86 | 52.61 | 84.33 | |
| VidAssistzero-shot=false2026.02 | 20.78 | 49.07 | 78.93 | |
| ActionDiffusion2026.02 | 18.04 | 44.54 | 56.23 | |
| LFP2026.02 | 15.97 | 50.7 | 75.3 | |
| KEPP2026.02 | 15.63 | 39.53 | 53.27 | |
| PDPP2026.02 | 14.41 | 44.1 | 51.39 | |
| E3P2026.02 | 13.59 | 26.72 | 84.72 | |
| P3IV2026.02 | 11.32 | 18.85 | 70.53 | |
| DDN2026.02 | 11.13 | 17.71 | 68.06 | |
| VidAssistzero-shot=true2026.02 | 9.07 | 42.72 | 80.83 | |
| Random2026.02 | 0.01 | 0.01 | 2.32 |