[docs] How to provide context to LLM RAG Evaluator when the documents are chunked and stored in a vector database

[railsstudent]

Description
I would like to use LLM RAG Evaluator to validate the response against the prompt and the context. This validator expects the context to be a string. However, my RAG application split the PDF document into chunked documents and store them in vector store. If I want to pass the context to the evaluator, I do not know how to do it. It is because qdrant returns the context in an array of JSON objects.

Please advice. Thanks.

Current documentation
https://hub.guardrailsai.com/validator/arize-ai/llm_rag_evaluator

Suggested changes
Please provide an example to provide the context to the validator when the documents are chunked and stored in a Qdrant vector database. The context is an array of JSON objects; do I concatenate the array to a string and assign the result to the context property of the metadata?

Additional context
[Add any other context, screenshots, or examples that could help explain the issue or improvement]

Checklist

• I have checked that this issue hasn't already been reported
• I have checked the latest version of the documentation to ensure this issue still exists
• For simple typos or fixes, I have considered submitting a pull request instead

[brianlai98]

Hi @railsstudent, yes that's correct, you would concatenate the contents of the retrieved chunks from Qdrant and pass that as the context field in the metadata.