Find answers from the community

Updated 4 months ago

Improving the Efficiency of BM25Retriever for PDF Documents

At a glance

The community member is working on building a BM25Retriever directly from PDF documents without chunking. They are concerned that if they add more documents to the system, they would have to rebuild the retriever from scratch on the entire set of documents. The community member is considering using qdrant's default hybrid search as a potential solution to this issue.

$$h0gun

One more thing I want to build this BM25Retriever from directly pdf documents without chunking at all and also If i add more documents to my system i have to build this retriever from scratch on the entire documents again, is there any better solution to this? I am working on hybrid search maybe qdrant's default hybrid search might help?

Add a reply