Long-Horizon Multi-Task Language Control on CALVIN ABC→D (test)
96Seq Success (1)CLOVER
Evaluation Results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| CLOVER2026.02 | 96 | 83.5 | 70.8 | 57.5 | 45.4 | 3.53 | |
| ReconVLA2026.02 | 95.6 | 87.6 | 76.9 | 69.3 | 64.1 | 3.95 | |
| UniVLA2026.02 | 95.5 | 85.8 | 75.4 | 66.9 | 56.5 | 3.8 | |
| VISTA-OFT2026.02 | 94.8 | 87.5 | 80.1 | 73.3 | 66.6 | 4.02 | |
| DITA2026.02 | 94.5 | 82.5 | 72.8 | 61.3 | 50 | 3.61 | |
| RoboDual2026.02 | 94.4 | 82.7 | 72.1 | 62.4 | 54.4 | 3.66 | |
| VISTA2026.02 | 93.5 | 78.6 | 64.6 | 53.9 | 43.1 | 3.34 | |
| RoboVLMsView=Single static third-person view2026.02 | 93.1 | 83.6 | 75.2 | 68.3 | 61.6 | 3.82 | |
| OpenVLAreproduced=true2026.02 | 92.8 | 78.3 | 63.7 | 52.3 | 44 | 3.31 | |
| OpenVLA-OFTreproduced=true2026.02 | 92.1 | 83.2 | 76.3 | 70.9 | 64.3 | 3.87 | |
| VPPView=Single static third-person view2026.02 | 90.9 | 81.5 | 71.3 | 62 | 51.8 | 3.58 | |
| SuSIE2026.02 | 87 | 69 | 49 | 38 | 26 | 2.69 | |
| GR-12026.02 | 85.4 | 71.2 | 59.6 | 49.7 | 40.1 | 3.06 |