Offline Multi-Agent Reinforcement Learning on Marine-Easy Medium
85.6Performance (3m Map)STAIRS-Former
Evaluation Results
| Method | Links | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| STAIRS-Former2026.03 | 85.6 | 85 | 94.4 | 73.8 | 82.5 | 98.1 | 96.9 | 93.1 | 65.6 | 65.6 | 84.1 | |
| HiSSD2026.03 | 67.5 | 80.6 | 66.3 | 73.8 | 86.3 | 93.8 | 90.6 | 76.9 | 48.1 | 40.6 | 72.5 | |
| ODIS2026.03 | 60 | 79.4 | 77.5 | 55 | 87.5 | 81.3 | 88.8 | 79.4 | 51.3 | 31.3 | 69.2 | |
| UPDeT-m2026.03 | 55.6 | 71.9 | 48.8 | 34.4 | 70.6 | 53.8 | 78.8 | 52.5 | 26.9 | 20 | 51.3 |