Cultural alignment on United States survey
64.86Soft SimilarityGPT-3.5
Evaluation results
| Method | Links | ||
|---|---|---|---|
| GPT-3.5Prompting Language=Averaged (English & Arabic)2024.02 | 64.86 | 39.29 | |
| LLaMA-2-ChatPrompting Language=Averaged (English & Arabic)2024.02 | 63.1 | 36.72 | |
| mT0-XXLPrompting Language=Averaged (English & Arabic)2024.02 | 55.48 | 31.4 | |
| AceGPT-ChatPrompting Language=Averaged (English & Arabic)2024.02 | 52.83 | 27.69 |