LLM Agent Planning on Main benchmark suite TextFrozenLake, CrafterMini, ALFWorld-A/B/C
100Aggregate Normalised Success RateLWM-Planner
Evaluation Results
| Method | Links | |
|---|---|---|
| LWM-Planner2025.06 | 100 | |
| ReAct + FEC2025.06 | 61.4 | |
| ReAct2025.06 | 48.1 | |
| Reflexion2025.06 | 43.1 | |
| Random2025.06 | 0 |