Embodied Task Execution on MEMENTO Acquisition Stage
16Planning CyclesClaude-3.5-Sonnet
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| Claude-3.5-SonnetStage=Acquisition, Task Type (Memory)=-2025.05 | 16 | 2,104.1 | 96.2 | 94 | |
| GPT-4oStage=Acquisition, Task Type (Memory)=-2025.05 | 16.5 | 2,156.1 | 96.3 | 95 | |
| Qwen-2.5-72bStage=Acquisition, Task Type (Memory)=-2025.05 | 17.5 | 2,281.9 | 93.5 | 91 | |
| Llama-3.1-70bStage=Acquisition, Task Type (Memory)=-2025.05 | 17.7 | 2,162.1 | 92.9 | 90 | |
| Llama-3.1-8bStage=Acquisition, Task Type (Memory)=-2025.05 | 19.3 | 2,377 | 78.1 | 68.5 | |
| Qwen-2.5-7bStage=Acquisition, Task Type (Memory)=-2025.05 | 21.7 | 2,476.8 | 64.1 | 53.2 |