Google

The Deliberate-then-Generate (DTG) prompt template for a Chinese-to-English translation task specifies the inputs by clearly labeling the source and target texts. The template uses `Given the Chinese sentence: {∗source∗}` to present the original sentence and `The English translation is: {∗target∗}` to present the initial, potentially flawed translation that requires refinement.

Input Structure of the DTG Prompt for Chinese-to-English Translation

A complete prompt template for the Deliberate-then-Generate (DTG) method in a Chinese-to-English translation task involves a multi-step structure. First, it presents the source Chinese sentence and its initial English translation. Then, it directs the model to analyze the translation by identifying any errors and subsequently generating a refined version. Finally, it cues the model's output by explicitly starting the response line with 'Error Type:'. The full template is structured as follows:

```
Given the Chinese sentence: {*source*}
The English translation is: {*target*}
Please first detect the type of error, and then refine the translation. 
Error Type:
```

Complete DTG Prompt Template for Chinese-to-English Translation

A user is constructing a prompt to have a language model assess and improve an existing translation from Chinese to English. The user's goal is to first present the necessary texts to the model before providing any specific instructions. Analyze the user's initial prompt snippet below. Identify the critical piece of information that is missing and explain why its absence makes the intended task impossible to perform correctly.

Critique of a Translation Evaluation Prompt

You are preparing a prompt to have a large language model refine an English translation of a Chinese sentence. Using the specific input format designed for this task, write the first two lines of the prompt. Use the following sentences:

- Chinese Source: `这道菜太咸了。`
- Initial English Translation: `This dish is too much salt.`

Constructing a Prompt Input for Translation Refinement

A user wants to prompt a large language model to first analyze and then improve an existing Chinese-to-English translation. Which of the following prompt snippets most effectively and neutrally establishes the initial context for this task, presenting the necessary information before any specific instructions are given?

Learn Before

Related