Find answers from the community

Updated 9 months ago

Rerank

can anyone understand why reranking makes the mean reciprocal rank worse than if i do not use reranking at all. I am using a hybrid retriever (BM25 n Index).
Attachment
image.png
L
b
3 comments
Maybe the reranker is just not great in this case?
I see Thanks, I assumed reranking would most certainly improve performance
I think it relly depends on the data the rerranker was trained on vs what data you are reranking
Add a reply
Sign up and join the conversation on Discord