English

Observations on Building RAG Systems for Technical Documents

Machine Learning 2024-04-02 v1 Artificial Intelligence Computation and Language

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.

Keywords

Cite

@article{arxiv.2404.00657,
  title  = {Observations on Building RAG Systems for Technical Documents},
  author = {Sumit Soman and Sujoy Roychowdhury},
  journal= {arXiv preprint arXiv:2404.00657},
  year   = {2024}
}

Comments

Published as a Tiny Paper at ICLR 2024

R2 v1 2026-06-28T15:39:33.076Z