RAG: Passing entire retrieved chunk as a string vs using only chunk.page_content

[bogdan-pechounov]

I just realized that the entire Document is passed in the context of the prompt, not just the document.page_content, is this intended? I was expecting the page content of each documents to be concatenated when inserted into the prompt.

from operator import itemgetter

from langchain_community.vectorstores import FAISS
from langchain_core.output_parsers import StrOutputParser
from langchain_core.prompts import ChatPromptTemplate
from langchain_core.runnables import RunnableLambda, RunnablePassthrough
from langchain_openai import ChatOpenAI, OpenAIEmbeddings

vectorstore = FAISS.from_texts(
    ["harrison worked at kensho", "harrison worked for 3 years"], embedding=OpenAIEmbeddings()
)
retriever = vectorstore.as_retriever()

template = """Answer the question based only on the following context:
{context}

Question: {question}
"""
prompt = ChatPromptTemplate.from_template(template)

model = ChatOpenAI()

chain = (
    {"context": retriever, "question": RunnablePassthrough()}
    | prompt
)

chain.invoke("where did harrison work?")

ChatPromptValue(messages=[HumanMessage(content="Answer the question based only on the following context:\n[Document(page_content='harrison worked for 3 years'), Document(page_content='harrison worked at kensho')]\n\nQuestion: where did harrison work?\n")])