Calendar Scheduling on Natural Plan Calendar Scheduling
72Success RateSMaRT
Evaluation Results
| Method | Links | |
|---|---|---|
| SMaRTModel=GPT-4, Number of shots=5-shot2025.10 | 72 | |
| CoTModel=GPT-4, Number of shots=5-shot2025.10 | 70.2 | |
| LLM-as-a-JudgeModel=GPT-4, Number of shots=5-shot2025.10 | 69.4 | |
| DirectModel=GPT-4, Number of shots=5-shot2025.10 | 66.1 | |
| SMaRTModel=Gemini-1.5, Number of shots=5-shot2025.10 | 58.5 | |
| CoTModel=Gemini-1.5, Number of shots=5-shot2025.10 | 57.7 | |
| LLM-as-a-JudgeModel=Gemini-1.5, Number of shots=5-shot2025.10 | 57 | |
| DirectModel=Gemini-1.5, Number of shots=5-shot2025.10 | 50.2 |