ValueError: The following model_kwargs are not used by the model: ['return_full_text'] when using RetrievalQA in LangChain

Question

I'm encountering a ValueError while using the RetrievalQA class in LangChain. The error message indicates that the model_kwargs parameter contains an unused parameter, 'return_full_text'. I've reviewed my code and ensured that this parameter is not being passed to the model initialization or the RetrievalQA instantiation.

qa = RetrievalQA.from_chain_type(
    llm=llm,
    chain_type="stuff",
    retriever=retriever,
    client_settings=CHROMA_SETTINGS
)

Despite removing the return_full_text parameter, the error persists. I suspect that the issue might be related to how the RetrievalQA class interacts with the underlying model.