Graph grounded retrieval improves answers on engineering standards
GEMS-RAG: Graph-grounded Engineering Multimodal Standards Retrieval-Augmented Generation
Artificial IntelligenceMachine Learning
Summary
Engineering standards are complicated because they include text, pictures, rules, and links all mixed together. The authors created GEMS-RAG, a system that breaks down both the text and images into connected pieces and uses these connections to find the best answers. By combining different searching methods and using a detailed graph of the whole document, their approach finds better information than previous methods. They tested it on traffic control manuals and showed it helps answer questions more accurately.
What this means in practice
- •For engineering document teams: Support faster and more accurate question answering over complex engineering manuals combining text and images.
- •For regulatory compliance teams: Help interpret multimodal regulatory codes and standards for improved compliance checks in safety-critical industries.
Authors
Abdul Hannan Azad, Irvin Wang, Jiwan Jiang, Chenchen Kuai, Bofeng Cao, Keshu Wu, Yuzhou Wang, Xiaopeng Li, Yang Zhou
Abstract
Engineering standards are challenging knowledge sources for large language models (LLMs) because they are hierarchical, multimodal, and densely connected through figures, tables, cross-references, and normative rules. Existing retrieval-augmented generation (RAG) methods often rely on unstructured text retrieval and struggle with questions requiring joint reasoning over textual provisions, visual specifications, and document structure. We propose GEMS-RAG (Graph-grounded Engineering Multimodal Standards Retrieval-Augmented Generation), a framework for grounded reasoning over engineering standards. GEMS-RAG combines hierarchical parsing, fine-grained visual entity extraction, multimodal knowledge graph construction, and graph-grounded hybrid retrieval. Instead of treating figures as monolithic images, it decomposes them into meaningful visual entities linked to textual provisions, engineering concepts, and regulatory rules. During inference, dense, sparse, and visual retrieval are combined with graph expansion to identify relevant evidence across text, figures, and document structure. We evaluate GEMS-RAG on the Manual on Uniform Traffic Control Devices (MUTCD), a large engineering standard containing egulations, technical drawings, and standardized traffic-control devices. Results show that graph-grounded multimodal retrieval improves retrieval effectiveness and question-answering performance over text-only, multimodal, and graph-based RAG baselines. Beyond MUTCD, GEMS-RAG offers a general framework for reasoning over engineering codes, technical manuals, regulatory documents, and scientific standards in safety-critical applications.