Embodied Task Completion on ALFWorld ID
94.3Pick Success RateSKILL0
Evaluation results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| SKILL0Category=Skill-Augmented RL Methods2026.05 | 94.3 | 76 | 81.5 | 85.1 | 3.8 | |
| Skill0.5Category=Skill-Augmented RL Methods2026.05 | 94.3 | 88 | 96.3 | 93.1 | 1.3 | |
| RLOOCategory=RL-based Methods2026.05 | 91.4 | 80 | 81.5 | 85.1 | 4.3 | |
| SkillRLCategory=Skill-Augmented RL Methods2026.05 | 91.4 | 84 | 96.3 | 90.8 | 2.7 | |
| SLIMCategory=Skill-Augmented RL Methods2026.05 | 91.4 | 84 | 70.4 | 82.8 | 4.5 | |
| EvolveRCategory=Memory-Augmented RL Methods2026.05 | 88.6 | 52 | 81.5 | 75.9 | 6.2 | |
| ReflexionCategory=Prompt-based Methods2026.05 | 85.7 | 44 | 44.4 | 60.9 | 10.3 | |
| SimpleMem+GRPOCategory=Memory-Augmented RL Methods2026.05 | 85.7 | 52 | 70.3 | 71.3 | 7.7 | |
| ExpeLCategory=Prompt-based Methods2026.05 | 80 | 44 | 66.7 | 65.5 | 9.5 | |
| GRPOCategory=RL-based Methods2026.05 | 80 | 72 | 88.9 | 80.5 | 6.2 | |
| MemRLCategory=Memory-Augmented RL Methods2026.05 | 74.3 | 12 | 55.6 | 50.6 | 12.7 | |
| ReActCategory=Prompt-based Methods2026.05 | 71.4 | 28 | 33.3 | 47.1 | 13.3 | |
| SimpleMemCategory=Prompt-based Methods2026.05 | 71.4 | 16 | 44.4 | 47.1 | 13.3 | |
| MemPCategory=Prompt-based Methods2026.05 | 65.7 | 12 | 33.3 | 40.2 | 15.3 | |
| Mem0+GRPOCategory=Memory-Augmented RL Methods2026.05 | 65.7 | 20 | 51.9 | 48.3 | 13.2 | |
| Few-shotCategory=Prompt-based Methods2026.05 | 62.9 | 44 | 63 | 57.5 | 12 | |
| Mem0Category=Prompt-based Methods2026.05 | 54.3 | 4 | 18.5 | 28.7 | 17.5 | |
| Zero-shotCategory=Prompt-based Methods2026.05 | 28.6 | 12 | 18.5 | 20.7 | 17.2 |