Techniques we want to look at: - [ ] adding more data during ingestion, for instance a summary of the document - [ ] [late chunking](https://jina.ai/news/late-chunking-in-long-context-embedding-models/) - [ ] [raptor](https://github.com/parthsarthi03/raptor)
Techniques we want to look at: