I’m working with LlamaIndex and I’m interested in implementing a Multishot Prompting mechanism. My goal is to use multiple prompts to refine and diversify the output for a given query.

Eg.:
You are going to translate technical text from English into Chinese using technical Chinese in the translation.

I will provide three examples of a technical paragraph in English and its corresponding translation in Chinese. Follow a similar structure when translating a new paragraph.

Paragraph 1: English Text

Translation: Chinese Translation

Paragraph 2: English Text

Translation: Chinese Translation

Paragraph 3: English Text

Translation: Chinese Translation

I tried it this way:

        examples = [
    {
        "question": "qestion",
        "answer": "answer",
    },
    {
        "question": "question",
        "answer": "answer",
    },
    {
        "question": "question",
        "answer": "answer",
    },
    {
        "question": "question",
        "answer": "Answer",
    },
    {
        "question": "question",
        "answer": "answer",
    },
    {
        "question": "question",
        "answer": "answer",
    },
]

text_qa_msgs = [
            ChatMessage(role=MessageRole.SYSTEM, content=system_message),
        ]
        for example in examples:
            text_qa_msgs.append(ChatMessage(role=MessageRole.USER, content=f"Q: {example['question']}nA: {example['answer']}"))
        text_qa_msgs.append(ChatMessage(role=MessageRole.USER, content=f"Q: {user_question}nA: "))
        text_qa_template = ChatPromptTemplate(text_qa_msgs)```

Is it even possible to build this into llama index?