| Benchmark | Model | Rank | Results |
|---|---|---|---|
| question-answering-on-fever | Self-Ask | #2 | EM: 64.2 |
| question-answering-on-bamboogle | Self-ask (GPT-3; davinci-002) + Google Search | #3 | Accuracy: 60.0 |
| question-answering-on-bamboogle | Self-ask (GPT-3; davinci-002) | #4 | Accuracy: 57.6 |
| question-answering-on-bamboogle | Chain-of-Thought (GPT-3; davinci-002) | #5 | Accuracy: 46.4 |
| question-answering-on-bamboogle | Direct Prompting (GPT-3; davinci-002) | #6 | Accuracy: 17.6 |
| question-answering-on-bamboogle | Google Search | #7 | Accuracy: 0 |
| question-answering-on-webquestions | Self-Ask | #20 | EM: 31.1 |