Creative Translation on CommonMT
70.7AccuracySelf-Contrast
Evaluation Results
| Method | Links | |
|---|---|---|
| Self-ContrastBackbone=GPT3.52024.01 | 70.7 | |
| Multi-AgentBackbone=GPT3.52024.01 | 69.9 | |
| ExpertPromptBackbone=GPT3.52024.01 | 69.6 | |
| Hint-PromptBackbone=GPT3.52024.01 | 69.6 | |
| Self-ReflectionBackbone=GPT3.52024.01 | 69.3 | |
| CoT PromptBackbone=GPT3.52024.01 | 69.1 | |
| SC-ReflectBackbone=GPT3.52024.01 | 69 | |
| SC-SelectBackbone=GPT3.52024.01 | 68.6 | |
| Self-ContrastBackbone=Llama2-70B-chat2024.01 | 66.7 | |
| Multi-AgentBackbone=Llama2-70B-chat2024.01 | 65.8 | |
| Hint-PromptBackbone=Llama2-70B-chat2024.01 | 64.6 | |
| ExpertPromptBackbone=Llama2-70B-chat2024.01 | 63.4 | |
| CoT PromptBackbone=Llama2-70B-chat2024.01 | 63.2 | |
| SC-ReflectBackbone=Llama2-70B-chat2024.01 | 63.2 | |
| Multi-AgentBackbone=Llama2-13B-chat2024.01 | 63.1 | |
| SC-SelectBackbone=Llama2-70B-chat2024.01 | 63 | |
| ExpertPromptBackbone=Llama2-13B-chat2024.01 | 62.9 | |
| SC-SelectBackbone=Llama2-13B-chat2024.01 | 62.8 | |
| Self-ContrastBackbone=Llama2-13B-chat2024.01 | 62.8 | |
| CoT PromptBackbone=Llama2-13B-chat2024.01 | 62.5 | |
| Hint-PromptBackbone=Llama2-13B-chat2024.01 | 62.5 | |
| Self-ReflectionBackbone=Llama2-70B-chat2024.01 | 62.2 | |
| SC-ReflectBackbone=Llama2-13B-chat2024.01 | 62.2 | |
| Self-ReflectionBackbone=Llama2-13B-chat2024.01 | 61.5 | |
| Hint-PromptBackbone=Llama2-7B-chat2024.01 | 54.2 | |
| SC-ReflectBackbone=Llama2-7B-chat2024.01 | 54 | |
| ExpertPromptBackbone=Llama2-7B-chat2024.01 | 53.8 | |
| CoT PromptBackbone=Llama2-7B-chat2024.01 | 53.7 | |
| SC-SelectBackbone=Llama2-7B-chat2024.01 | 52.1 | |
| Self-ContrastBackbone=Llama2-7B-chat2024.01 | 52.1 | |
| Multi-AgentBackbone=Llama2-7B-chat2024.01 | 51.9 | |
| Self-ReflectionBackbone=Llama2-7B-chat2024.01 | 48.8 |