Embodied Planning on MRoom-30k 1.0 (test)
56.94SimilarityOOWM 3-Stage
Evaluation Results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| OOWM 3-StagePipeline Configuration=Full Pipeline2026.02 | 56.94 | 53.26 | 67.44 | 67.44 | 59.04 | |
| OOWM 2-StagePipeline Configuration=OOWM → OOWM2026.02 | 56.17 | 53.04 | 65.36 | 65.36 | 58.03 | |
| Hybrid StrategyPipeline Configuration=Text → OOWM2026.02 | 55.62 | 53.84 | 64.38 | 64.38 | 58.12 | |
| Unstructured BaselineReasoning Strategy=Text → Text2026.02 | 54.98 | 54.89 | 62.8 | 62.8 | 58.11 | |
| Graph of ThoughtsType=Baseline2026.02 | 53.83 | 52.63 | 55.79 | 55.79 | 53.71 | |
| Tree of ThoughtsType=Baseline2026.02 | 42.09 | 48.54 | 46.39 | 46.39 | 46.95 |