Sequential Decision Making on ALFWorld (134 evaluation tasks)
71.5Normalized Success RateLWM-Planner
Evaluation Results
| Method | Links | |
|---|---|---|
| LWM-Plannerruns=single-seed, confidence_intervals=95%2025.06 | 71.5 | |
| ReAct + FECruns=single-seed, confidence_intervals=95%2025.06 | 68.1 | |
| ReActruns=single-seed, confidence_intervals=95%2025.06 | 46.9 | |
| Reflexionruns=single-seed, confidence_intervals=95%2025.06 | 37.4 | |
| Randomruns=single-seed, confidence_intervals=95%2025.06 | 0 |