| Dataset Name | SOTA Method | Metric | Trend | ||
|---|---|---|---|---|---|
| ALFWorld held-out (test) | SkillOpt | Score95.5 | 49 | 9d ago | |
| ALFWorld | UMEM-Qwen3-4B | Success Rate82.84 | 31 | 3mo ago | |
| ALFWorld (ID and OOD) | RL w/ ACT | Success Rate92.86 | 16 | 2mo ago | |
| Game for Peace 1.0 (test) | MAIN-VLA | Parachuting Score71.4 | 6 | 3mo ago |