Anyone is using Qdrant with LlamaIndex? I got frustrated because it keeps deleting my collection hence makes us data loss because it keeps trigerring the recreate_collection function. I found the same issue raised by this guy (https://github.com/run-llama/llama_index/issues/9352) and seems like it's still haven't fixed yet? I assure that I use llama index from back then too (before Dec 2023) and tbh I haven't updated our LlamaIndex yet.

Hello, I want to ask. If we make a retrieval using BM25, then add the documents that we want to be retrieved, so we have to re-index from the beginning, right? Because I think the Document Frequency has changed, right? CMIIW