屬於 [[Query Optimization 最佳化策略]] 產生多個相似 queries > Given a single user query, use an LLM to synthetically generate multiple other queries. Use each one of the new queries to retrieve documents, take the union of those documents for the final context of your prompt https://twitter.com/dotey/status/1718144505226940605 * langchain 的 multiquery * https://python.langchain.com/docs/modules/data_connection/retrievers/MultiQueryRetriever Q: multiquery 拉出來的資料，是怎麼融合排序的????? 也是用 ensemble 嗎? prompt: You are an AI language model assistant. Your task is to generate five different versions of the given user question to retrieve relevant documents from a vector database. By generating multiple perspectives on the user question, your goal is to help the user overcome some of the limitations of the distance-based similarity search. Provide these alternative questions separated by newlines. Original question: {question} * https://medium.com/@kbdhunga/advanced-rag-multi-query-retriever-approach-ad8cd0ea0f5b ## Reciprocal Rank Fusion 合併法 Multi-Queries 常搭配 Fusion 合併法，也就是 [[Ensemble Retriever]] https://plg.uwaterloo.ca/~gvcormac/cormacksigir09-rrf.pdf https://github.com/Raudaschl/rag-fusion https://twitter.com/jerryjliu0/status/1713573483228356733 https://towardsdatascience.com/forget-rag-the-future-is-rag-fusion-1147298d8ad1 https://github.com/langchain-ai/langchain/blob/master/cookbook/rag_fusion.ipynb?ref=blog.langchain.dev * 可以用LLM產生多個相似 query，每個 query 再用 向量搜尋.... * https://github.com/Raudaschl/rag-fusion/tree/master * https://docs.llamaindex.ai/en/latest/examples/low_level/fusion_retriever.html