Multi-Agent Reinforcement Learning on SMAC Stalker-Zealot Medium-Replay quality (Source Tasks)
20.6Performance (2s3z)STAIRS-Former
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| STAIRS-Former2026.03 | 20.6 | 28.8 | 28.8 | 24.3 | |
| ODIS2026.03 | 10.6 | 6.3 | 12.5 | 5 | |
| HiSSD2026.03 | 5.6 | 7.5 | 22.5 | 15 | |
| UPDeT-m2026.03 | 3.8 | 10 | 5 | 3.6 |