How to use the MultiVector Retriever

It can often be beneficial to store multiple vectors per document. There are multiple use cases where this is beneficial. LangChain has a base MultiVectorRetriever which makes querying this type of setup easy. A lot of the complexity lies in how to create the multiple vectors per document. This notebook covers some of the common ways to create those vectors and use the MultiVectorRetriever.

The methods to create multiple vectors per document include:

• Smaller chunks: split a document into smaller chunks, and embed those (this is ParentDocumentRetriever).
• Summary: create a summary for each document, embed that along with (or instead of) the document.
• Hypothetical questions: create hypothetical questions that each document would be appropriate to answer, embed those along with (or instead of) the document.

Note that this also enables another method of adding embeddings - manually. This is great because you can explicitly add questions or queries that should lead to a document being recovered, giving you more control.

from langchain.retrievers.multi_vector import MultiVectorRetriever

API Reference:MultiVectorRetriever

from langchain.storage import InMemoryByteStore
from langchain_chroma import Chroma
from langchain_community.document_loaders import TextLoader
from langchain_openai import OpenAIEmbeddings
from langchain_text_splitters import RecursiveCharacterTextSplitter

API Reference:InMemoryByteStore | TextLoader | OpenAIEmbeddings | RecursiveCharacterTextSplitter

loaders = [
    TextLoader("../../paul_graham_essay.txt"),
    TextLoader("state_of_the_union.txt"),
]
docs = []
for loader in loaders:
    docs.extend(loader.load())
text_splitter = RecursiveCharacterTextSplitter(chunk_size=10000)
docs = text_splitter.split_documents(docs)

Smaller chunks

Often times it can be useful to retrieve larger chunks of information, but embed smaller chunks. This allows for embeddings to capture the semantic meaning as closely as possible, but for as much context as possible to be passed downstream. Note that this is what the ParentDocumentRetriever does. Here we show what is going on under the hood.

# The vectorstore to use to index the child chunks
vectorstore = Chroma(
    collection_name="full_documents", embedding_function=OpenAIEmbeddings()
)
# The storage layer for the parent documents
store = InMemoryByteStore()
id_key = "doc_id"
# The retriever (empty to start)
retriever = MultiVectorRetriever(
    vectorstore=vectorstore,
    byte_store=store,
    id_key=id_key,
)
import uuid

doc_ids = [str(uuid.uuid4()) for _ in docs]

# The splitter to use to create smaller chunks
child_text_splitter = RecursiveCharacterTextSplitter(chunk_size=400)

sub_docs = []
for i, doc in enumerate(docs):
    _id = doc_ids[i]
    _sub_docs = child_text_splitter.split_documents([doc])
    for _doc in _sub_docs:
        _doc.metadata[id_key] = _id
    sub_docs.extend(_sub_docs)

retriever.vectorstore.add_documents(sub_docs)
retriever.docstore.mset(list(zip(doc_ids, docs)))

# Vectorstore alone retrieves the small chunks
retriever.vectorstore.similarity_search("justice breyer")[0]

Document(page_content='Tonight, I’d like to honor someone who has dedicated his life to serve this country: Justice Stephen Breyer—an Army veteran, Constitutional scholar, and retiring Justice of the United States Supreme Court. Justice Breyer, thank you for your service. \n\nOne of the most serious constitutional responsibilities a President has is nominating someone to serve on the United States Supreme Court.', metadata={'doc_id': '2fd77862-9ed5-4fad-bf76-e487b747b333', 'source': 'state_of_the_union.txt'})

# Retriever returns larger chunks
len(retriever.get_relevant_documents("justice breyer")[0].page_content)

9875

The default search type the retriever performs on the vector database is a similarity search. LangChain Vector Stores also support searching via Max Marginal Relevance so if you want this instead you can just set the search_type property as follows:

from langchain.retrievers.multi_vector import SearchType

retriever.search_type = SearchType.mmr

len(retriever.get_relevant_documents("justice breyer")[0].page_content)

API Reference:SearchType

9875

Summary

Oftentimes a summary may be able to distill more accurately what a chunk is about, leading to better retrieval. Here we show how to create summaries, and then embed those.

import uuid

from langchain_core.documents import Document
from langchain_core.output_parsers import StrOutputParser
from langchain_core.prompts import ChatPromptTemplate
from langchain_openai import ChatOpenAI

API Reference:Document | StrOutputParser | ChatPromptTemplate | ChatOpenAI

chain = (
    {"doc": lambda x: x.page_content}
    | ChatPromptTemplate.from_template("Summarize the following document:\n\n{doc}")
    | ChatOpenAI(max_retries=0)
    | StrOutputParser()
)

summaries = chain.batch(docs, {"max_concurrency": 5})

# The vectorstore to use to index the child chunks
vectorstore = Chroma(collection_name="summaries", embedding_function=OpenAIEmbeddings())
# The storage layer for the parent documents
store = InMemoryByteStore()
id_key = "doc_id"
# The retriever (empty to start)
retriever = MultiVectorRetriever(
    vectorstore=vectorstore,
    byte_store=store,
    id_key=id_key,
)
doc_ids = [str(uuid.uuid4()) for _ in docs]

summary_docs = [
    Document(page_content=s, metadata={id_key: doc_ids[i]})
    for i, s in enumerate(summaries)
]

retriever.vectorstore.add_documents(summary_docs)
retriever.docstore.mset(list(zip(doc_ids, docs)))

# # We can also add the original chunks to the vectorstore if we so want
# for i, doc in enumerate(docs):
#     doc.metadata[id_key] = doc_ids[i]
# retriever.vectorstore.add_documents(docs)

sub_docs = vectorstore.similarity_search("justice breyer")

sub_docs[0]

Document(page_content="The document is a speech given by President Biden addressing various issues and outlining his agenda for the nation. He highlights the importance of nominating a Supreme Court justice and introduces his nominee, Judge Ketanji Brown Jackson. He emphasizes the need to secure the border and reform the immigration system, including providing a pathway to citizenship for Dreamers and essential workers. The President also discusses the protection of women's rights, including access to healthcare and the right to choose. He calls for the passage of the Equality Act to protect LGBTQ+ rights. Additionally, President Biden discusses the need to address the opioid epidemic, improve mental health services, support veterans, and fight against cancer. He expresses optimism for the future of America and the strength of the American people.", metadata={'doc_id': '56345bff-3ead-418c-a4ff-dff203f77474'})

retrieved_docs = retriever.get_relevant_documents("justice breyer")

len(retrieved_docs[0].page_content)

9194

Hypothetical Queries

An LLM can also be used to generate a list of hypothetical questions that could be asked of a particular document. These questions can then be embedded

functions = [
    {
        "name": "hypothetical_questions",
        "description": "Generate hypothetical questions",
        "parameters": {
            "type": "object",
            "properties": {
                "questions": {
                    "type": "array",
                    "items": {"type": "string"},
                },
            },
            "required": ["questions"],
        },
    }
]

from langchain_core.output_parsers.openai_functions import JsonKeyOutputFunctionsParser

chain = (
    {"doc": lambda x: x.page_content}
    # Only asking for 3 hypothetical questions, but this could be adjusted
    | ChatPromptTemplate.from_template(
        "Generate a list of exactly 3 hypothetical questions that the below document could be used to answer:\n\n{doc}"
    )
    | ChatOpenAI(max_retries=0, model="gpt-4").bind(
        functions=functions, function_call={"name": "hypothetical_questions"}
    )
    | JsonKeyOutputFunctionsParser(key_name="questions")
)

API Reference:JsonKeyOutputFunctionsParser

chain.invoke(docs[0])

["What was the author's first experience with programming like?",
 'Why did the author switch their focus from AI to Lisp during their graduate studies?',
 'What led the author to contemplate a career in art instead of computer science?']

hypothetical_questions = chain.batch(docs, {"max_concurrency": 5})

# The vectorstore to use to index the child chunks
vectorstore = Chroma(
    collection_name="hypo-questions", embedding_function=OpenAIEmbeddings()
)
# The storage layer for the parent documents
store = InMemoryByteStore()
id_key = "doc_id"
# The retriever (empty to start)
retriever = MultiVectorRetriever(
    vectorstore=vectorstore,
    byte_store=store,
    id_key=id_key,
)
doc_ids = [str(uuid.uuid4()) for _ in docs]

question_docs = []
for i, question_list in enumerate(hypothetical_questions):
    question_docs.extend(
        [Document(page_content=s, metadata={id_key: doc_ids[i]}) for s in question_list]
    )

retriever.vectorstore.add_documents(question_docs)
retriever.docstore.mset(list(zip(doc_ids, docs)))

sub_docs = vectorstore.similarity_search("justice breyer")

sub_docs

[Document(page_content='Who has been nominated to serve on the United States Supreme Court?', metadata={'doc_id': '0b3a349e-c936-4e77-9c40-0a39fc3e07f0'}),
 Document(page_content="What was the context and content of Robert Morris' advice to the document's author in 2010?", metadata={'doc_id': 'b2b2cdca-988a-4af1-ba47-46170770bc8c'}),
 Document(page_content='How did personal circumstances influence the decision to pass on the leadership of Y Combinator?', metadata={'doc_id': 'b2b2cdca-988a-4af1-ba47-46170770bc8c'}),
 Document(page_content='What were the reasons for the author leaving Yahoo in the summer of 1999?', metadata={'doc_id': 'ce4f4981-ca60-4f56-86f0-89466de62325'})]

retrieved_docs = retriever.get_relevant_documents("justice breyer")

len(retrieved_docs[0].page_content)

9194