Few-shot task expansion on ALFWorld unseen domains
71.58SRTMoW
Evaluation results
| Method | Links | ||
|---|---|---|---|
| TMoWshot=average2026.01 | 71.58 | 29.98 | |
| SayCanPayshot=average2026.01 | 47.71 | 38.28 | |
| LLM+FTshot=average2026.01 | 45.99 | 37.66 | |
| FLAREshot=average2026.01 | 11.87 | 44.04 | |
| LLM-Plannershot=average2026.01 | 8.18 | 47.11 |