Design about TextSplitter - allow multiple RAG techniques
Nobody has claimed this yet.
- Dominant language
- C#
- Stars
- 1.1k
- Forks
- 141
- Avg merge
- 2d 18m
- Merged PRs (30d)
- 2
Description
I was looking if there is support for hierarchical chunks similar with GraphRAG and I see not there is not yet support.
Now there are many RAG techniques and using string as a chunk, I think it limits the library very much.
What I want to do is to create Contextual Retrieval but also be able to add metadata in a chunk from heading.
I have an idea how this can be improved, but first: do you see as a priority?
Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
Start by reviewing the existing TextSplitter abstraction and how chunks are currently represented as strings. Check whether heading metadata or hierarchical chunks are supported, then document the needed design for contextual retrieval and multiple RAG techniques. Done requires an agreed scope and implementation plan before coding.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- csharp
- Domain
- ai
- Issue type
- Feature
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 25/100