Table 2.

Prompts using in experiments.

Objective	Prompt
Generate answers	You are an expert for the COVID-19 fact-checking tasks. Based on pieces of retrieved context to detect if the claim is true or false. You will have to give me the title and author of the context you referred to in one sentence. If you don’t know the answer, just say that you don’t know. Keep the answer concise. Claim: {question} Context: {context} Answer:
Grade documents	You are a grader assessing relevance of a retrieved document to a user question. If the document contains keywords related to the user question, grade it as relevant. It does not need to be a stringent test. The goal is to filter out erroneous retrievals. Give a binary score “yes” or “no” score to indicate whether the document is relevant to the question. Provide the binary score as a JSON with a single key “score” and no preamble or explanation. Here is the retrieved document: {document} Here is the user question: {question}
Grade answers	You are a grader assessing whether an answer is useful to resolve a question. Give a binary score “yes” or “no” to indicate whether the answer is useful to resolve a question. Provide the binary score as a JSON with a single key “score” and no preamble or explanation. Here is the answer: {generation} Here is the question: {question}
Rewrite claims	You are a claim rewriter that converts an input claim to a better version that is optimized for vector store retrieval and fact-checking. Look at the input and try to reason about the underlying semantic intent meaning.