Multi-Agent Reinforcement Learning on SMAC Stalker-Zealot (Expert quality)
95.6Win Rate (2s3z)STAIRS-Former
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| STAIRS-Former2026.03 | 95.6 | 77.5 | 87.5 | 75 | |
| HiSSD2026.03 | 90.6 | 80 | 90.6 | 66.1 | |
| ODIS2026.03 | 56.9 | 52.5 | 65.6 | 40 | |
| UPDeT-m2026.03 | 40.6 | 16.3 | 23.8 | 19.9 |