How do you go about refreshing pinecone data? The data will come from Google Docs files. Do I delete the data and re-upload it?

When querying an index, how do I define that I only want to return data from 1 source?

When querying your index via "query_engine.query()", is it possible to add source to the response? Do I need to grab it via "response.get_formatted_sources()" ?

I'm prompting a private KB and I asked a question that doesn't have the answer in the KB. The query still found 1 node with 0.74 similarity and then made up a response which has nothing to do with the question. How do I prevent this from happening? I'm using GPTVectorStoreIndex to generate the index and my query engine is index.as_query_engine(similarity_top_k=1, retriever_mode="embedding"). Model name is "text-ada-001"