top_k behavior in GraphCypherQAChain

[minyunhong]

Checked other resources

• I added a very descriptive title to this question.
• I searched the LangChain documentation with the integrated search.
• I used the GitHub search to find a similar question and didn't find it.

Commit to Help

• I commit to help with one of those options 👆

Example Code

# 1. from langchain_neo4j import GraphCypherQAChain
# -> langchain_neo4j: v0.4.0
# https://python.langchain.com/api_reference/neo4j/chains/langchain_neo4j.chains.graph_qa.cypher.GraphCypherQAChain.html
# 2. from langchain.chains import GraphCypherQAChain
# -> langchain: v0.3.7 or v0.3.20
# https://api.python.langchain.com/en/latest/chains/langchain.chains.graph_qa.cypher.GraphCypherQAChain.html
# 3. from langchain_community.chains.graph_qa.cypher import GraphCypherQAChain
# -> langchain_community: v0.3.7 or v0.3.20
# https://python.langchain.com/api_reference/community/chains/langchain_community.chains.graph_qa.cypher.GraphCypherQAChain.html

    chain = GraphCypherQAChain.from_llm(
        cypher_llm = cypher_llm,
        qa_llm = qa_llm,
        graph = graph,
        cypher_prompt = cypher_prompt,
        qa_prompt = qa_prompt,
        verbose = True,
        validate_cypher = True,
        top_k = 11,
        return_intermediate_steps = True,
        allow_dangerous_requests = True
    )

Description

Hi,

I expect top_k to return multiple responses, whether in Cypher queries or QA responses. However, I only receive a single Cypher query and a single QA response.

Is there anything additional that needs to be configured for top_k to function correctly?
(It does not seem to work even when modifying the prompt. Interestingly, when combining Cypher queries and prompts using langchain.chains.llm.LLMChain, I can retrieve multiple results as expected.)
Since top_k is an existing parameter, I would like to know how to obtain multiple responses using top_k alone without additional modifications.

I appreciate your insights on this!

Thanks.

System Info

---

OS: Linux
OS Version: #66~20.04.1-Ubuntu SMP Wed Jan 25 09:41:30 UTC 2023
Python Version: 3.12.9 (main, Feb 4 2025, 14:38:38) [GCC 11.4.0]

Package Information

langchain_core: 0.3.31
langchain: 0.3.7
langchain_community: 0.3.7
langsmith: 0.1.143
langchain_google_genai: 2.0.7
langchain_groq: 0.2.1
langchain_neo4j: 0.4.0
langchain_openai: 0.3.2
langchain_text_splitters: 0.3.2

or

langchain: 0.3.20
langchain_community: 0.3.20