Symbolic Reasoning on Coinflip 4
99.5AccuracySelf-consistency
Evaluation Results
| Method | Links | |
|---|---|---|
| Self-consistencyModel=GPT-3 (Code-davinci-002)2022.03 | 99.5 | |
| CoT-promptingModel=GPT-3 (Code-davinci-002)2022.03 | 99 | |
| Self-consistencyModel=PaLM-540B2022.03 | 91.2 | |
| CoT-promptingModel=PaLM-540B2022.03 | 88.2 | |
| Self-consistencyModel=GPT-3 (Code-davinci-001)2022.03 | 75.9 | |
| Self-consistencyModel=LaMDA-137B2022.03 | 73.5 | |
| CoT-promptingModel=LaMDA-137B2022.03 | 72.4 | |
| CoT-promptingModel=GPT-3 (Code-davinci-001)2022.03 | 71.4 | |
| Self-consistencyModel=UL2-20B2022.03 | 50.5 | |
| CoT-promptingModel=UL2-20B2022.03 | 50.4 |