Multi-Robot Collaboration on RoCoBench Move Rope
86Task Success RateREFLEX + GPT-4
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| REFLEX + GPT-4Framework=REFLEX, Base LLM=GPT-42025.05 | 86 | 2.1 | 1.2 | |
| REFLEX + LLaMA-3.1Framework=REFLEX, Base LLM=LLaMA-3.12025.05 | 76 | 2 | 2.4 | |
| RoCo + GPT-4Framework=RoCo, Base LLM=GPT-42025.05 | 65 | 2.5 | 3.1 | |
| Central PlanFramework=Central Plan2025.05 | 50 | 2.3 | 3.9 |