High-level Planning on ALFRED (val unseen)
64EMAlfred
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| AlfredModel=Vicuna, Size=7B, N.Tasks=812, Grounding=Yes2025.12 | 64 | 86 | 78 | |
| AlfredModel=Vicuna, Size=7B, N.Tasks=812, Grounding=No2025.12 | 35 | 64 | 59 | |
| InnerMonologueModel=Vicuna, Size=7B, N.Tasks=812, Grounding=Yes2025.12 | 15 | 50 | 38 | |
| ProgPromptModel=Vicuna, Size=7B, N.Tasks=812, Grounding=No2025.12 | 12 | 26 | 20 |