Interactive Planning on ALFRED unseen (val)
67.8Success Rate (SR)Qwen2-VL-72B
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| Qwen2-VL-72BModel Parameters=72B2024.09 | 67.8 | 75.8 | — | |
| Previous SOTA2024.09 | 67.7 | 75.3 | — | |
| Scene-LLMStep-by-step instructions=false2024.03 | 22.35 | 38.86 | 79.63 | |
| HLSMStep-by-step instructions=false2024.03 | 18.28 | 31.24 | 70.17 | |
| LLM-PlannerStep-by-step instructions=true2024.03 | 15.36 | 29.88 | 68.31 | |
| LLM-PlannerStep-by-step instructions=false2024.03 | 12.92 | 25.35 | 55.85 | |
| HITUTStep-by-step instructions=false2024.03 | 10.23 | 20.71 | — | |
| E.T.Step-by-step instructions=true2024.03 | 7.32 | 20.87 | — | |
| FILMStep-by-step instructions=true2024.03 | — | — | 54.93 |