RAG over research papers
Dense, BM25 and hybrid retrieval benchmarked on a hand-labelled question set. Hybrid lifted chunk-level Hit@5 from 0.66 to 0.78.
- Type
- Build
- Tracks
- LLMs
- Context
- LangChain · FAISS · Chroma
- Links
- Code ↗
FIG. 10FIG. 10FIG. 10Dense, BM25 and hybrid retrieval benchmarked on a hand-labelled question set. Hybrid lifted chunk-level Hit@5 from 0.66 to 0.78.
FIG. 10