Easily use and train state of the art late-interaction retrieval methods (ColBERT) in any RAG pipeline. Designed for modularity and ease-of-use, backed by research.
Why does the function the RAG.rerank function does not use an index to rerank the documents?
For example, in the original paper of "colBERT" they used an already indexed documents representations in the "Top-𝑘 Re-ranking' part.
It seems more eficient and accurate to the paper.
Thanks!
Why does the function the RAG.rerank function does not use an index to rerank the documents? For example, in the original paper of "colBERT" they used an already indexed documents representations in the "Top-𝑘 Re-ranking' part. It seems more eficient and accurate to the paper. Thanks!