Multi-task Reinforcement Learning on Meta-World MT50 V2
90.9Overall Success Rate
TOPPO
Updated 21d ago
Evaluation Results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
2026.05 | 90.9 | 24.2 | 56.5 | 77.2 | 84.8 | 88.6 | |
2026.05 | 90.1 | 22.3 | 52.4 | 75.1 | 83.4 | 87.6 | |
2026.05 | 88.9 | - | - | - | - | - | |
2026.05 | 88.9 | 18 | 48.9 | 72.3 | 81.5 | 86.1 | |
2026.05 | 87.5 | 6.9 | 42.3 | 68.9 | 79.2 | 84.4 | |
2026.05 | 83.2 | - | 29.3 | 58.4 | 72 | 79 | |
2026.05 | 78.5 | 0 | 6.6 | 46.3 | 64.1 | 73.1 | |
2026.05 | 78.3 | - | 21 | 49.1 | 64.3 | 72.9 | |
2026.05 | 65.9 | - | 9.1 | 26.3 | 45.3 | 57.5 | |
2026.05 | 64.2 | 0 | 0 | 15.9 | 41.2 | 55.3 | |
2026.05 | 56 | 0 | 0 | 8 | 26.8 | 45 | |
2026.05 | 55.6 | 0 | 0 | 4.6 | 26.1 | 44.6 | |
2026.05 | 54.1 | 0 | 0 | 1.8 | 23.7 | 42.6 | |
2026.05 | 51.9 | 0 | 0 | 0 | 21 | 39.9 | |
2026.05 | 47.6 | 0 | 0 | 0 | 14.1 | 34.5 | |
2026.05 | 47.2 | 0 | 0 | 0 | 14.1 | 34 |