# Retrieval-Augmented Generation for Scientific Code Understanding

arXiv: https://arxiv.org/abs/2609.12190 · HTML: https://mlsanity.com/paper/2609.12190/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in RAG for code, one of 15 directions in Retrieval-augmented generation.
- Builds on 9 of its 28 references that are in the map.
- First posted on arXiv 10 Sep 2026.

## Placed in

- Retrieval-augmented generation / [RAG for code](https://mlsanity.com/rag/rag-for-code/)

## Builds on (most cited first, 9 of 9)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks](https://mlsanity.com/paper/2005.11401/) (2020)
- [Judging LLM-as-a-Judge with MT-Bench and Chatbot Arena](https://mlsanity.com/paper/2306.05685/) (2023)
- [Qwen2.5 Technical Report](https://mlsanity.com/paper/2412.15115/) (2024)
- [Billion-scale similarity search with GPUs](https://mlsanity.com/paper/1702.08734/) (2017)
- [RepoCoder: Repository-Level Code Completion Through Iterative Retrieval and Generation](https://mlsanity.com/paper/2303.12570/) (2023)
- [Nomic Embed: Training a Reproducible Long Context Text Embedder](https://mlsanity.com/paper/2402.01613/) (2024)
- [Gemma 4 Technical Report](https://mlsanity.com/paper/2607.02770/) (2026)
- [Scaling and performance portability of the particle-in-cell scheme for plasma physics applications through mini-apps targeting exascale architectures](https://mlsanity.com/paper/2205.11052/) (2022)

## Abstract (arXiv)

Large language models have become central to modern coding assistants, but state-of-the-art systems such as Claude Code or Codex rely on very large, cloud-hosted models with significant computational cost and data-privacy implications. This work investigates whether a useful, fully local coding agent can be built around small open-source models by shifting the computational burden away from inference. We develop a Retrieval-Augmented Generation (RAG) system for scientific code understanding that strictly separates an expensive offline ingestion stage parsing, structural graph construction, LLM-generated entity explanations, and embedding from a lightweight online answering stage. The system is evaluated on a 100-question benchmark spanning eleven categories over the IPPL scientific codebase written in C++, with answers scored by an independent frontier model as the judge. Across seven answering models, we find that model family and retrieval quality matter more than parameter count, i.e. a 9B model achieves the highest average score (0.795), outperforming both larger models within our pipeline and the same models embedded in the Claude Code retrieval architecture. The results indicate that front-loading code understanding into a reusable, codebase-specialised vector store enables small local models to deliver grounded and repository-specific answers, making the agent well suited as a privacy-preserving development tool for in-house scientific codebases.