Support for RAG Fusion


I’d like to request support for RAG Fusion. Essentially, it addresses a limitation in RAG by generating multiple queries and reranking the results (using Reciprocal Rank Fusion and custom vector score weighting).

Here’s an article that explains it in detail: Forget RAG, the Future is RAG-Fusion | by Adrian H. Raudaschl | Towards Data Science


Peter Cotroneo

I believe this can be implemented outside of pinecone as well.