Robot Task Planning on REI-Bench Implicit REs
40.1Object Omission RateLLaMA3.1-8B + TOCC
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| LLaMA3.1-8B + TOCCBase Model=LLaMA3.1-8B, Context Condition=Standard Context2025.05 | 40.1 | 30.6 | 70.7 | |
| LLaMA3.1-8B + CoTBase Model=LLaMA3.1-8B, Context Condition=Standard Context2025.05 | 47.6 | 30.3 | 77.9 | |
| LLaMA3.1-8B + APBase Model=LLaMA3.1-8B, Context Condition=Standard Context2025.05 | 49.9 | 27.4 | 77.3 | |
| LLaMA3.1-8B + ICLBase Model=LLaMA3.1-8B, Context Condition=Standard Context2025.05 | 49.9 | 28.7 | 78.6 | |
| LLaMA3.1-8BBase Model=LLaMA3.1-8B, Context Condition=Standard Context2025.05 | 53.9 | 24 | 77.9 | |
| LLaMA3.1-8B - ContextBase Model=LLaMA3.1-8B, Context Condition=Standard Context2025.05 | 85.1 | 5.5 | 90.6 |