Average Performance across 10 Task Types on 13 Datasets (test)
75.8Avg. AccuracyPROMPTED
Evaluation Results
| Method | Links | |
|---|---|---|
| PROMPTEDModel=GPT-42023.10 | 75.8 | |
| PROMPTEDModel=GPT-3.52023.10 | 68.8 | |
| Output RefinementModel=GPT-42023.10 | 68.6 | |
| Zero-Shot CoTModel=GPT-42023.10 | 67.3 | |
| Zero-ShotModel=GPT-42023.10 | 65.7 | |
| Output RefinementModel=GPT-3.52023.10 | 64.1 | |
| Zero-Shot CoTModel=GPT-3.52023.10 | 63.4 | |
| Zero-ShotModel=GPT-3.52023.10 | 62.2 |