Observations on Building RAG Systems for Technical Documents

arXiv:2404.00657v1 Announce Type: new
Abstract: Retrieval augmented generation (RAG) for technical documents creates challenges as embeddings do not often capture domain information. We review prior art for important factors affecting RAG and perform experiments to highlight best practices and potential challenges to build RAG systems for technical documents.

abstract art arxiv best practices build building challenges cs.ai cs.cl cs.lg documents domain embeddings highlight information practices prior rag retrieval retrieval augmented generation review systems technical type