Tabular Reasoning on INFOTABS (dev)
84.83AccuracyRethinking with retrieval
Evaluation results
| Method | Links | |
|---|---|---|
| Rethinking with retrievalModel=GPT-3 (text-davinci-002), Temperature=0.7, Num samples=102022.12 | 84.83 | |
| Self-consistencyModel=GPT-3 (text-davinci-002), Temperature=0.7, Num samples=102022.12 | 84 | |
| Chain-of-thought promptingModel=GPT-3 (text-davinci-002), Temperature=02022.12 | 83.33 | |
| Few-shot promptingModel=GPT-3 (text-davinci-002), Temperature=02022.12 | 83.08 | |
| Zero-shot promptingModel=GPT-3 (text-davinci-002), Temperature=02022.12 | 82 |