Planning on MAT2-THOR Overall
12.83Planning Time (s)LLM as a Planner
Evaluation results
| Method | Links | |
|---|---|---|
| LLM as a PlannerLLM Backbone=GPT-5.22026.03 | 12.83 | |
| LLM+P (PDDL based)LLM Backbone=GPT-5.22026.03 | 22.32 | |
| LaMMA-P (PDDL plan only)LLM Backbone=GPT-5.22026.03 | 57.59 | |
| Scale PlanLLM Backbone=GPT-5.22026.03 | 62.27 | |
| LaMMA-P (LLM corrected)LLM Backbone=GPT-5.22026.03 | 64.39 |