SRAG: RAG with Structured Data Improves Vector Retrieval

TL;DR

SRAG enhances vector retrieval for LLMs by incorporating structured data like topics, sentiments, and knowledge graph triples, significantly improving answer quality in question answering systems.

Contribution

The paper introduces Structured RAG (SRAG), a novel method that adds structured information to improve retrieval accuracy in RAG systems.

Findings

30% improvement in answer scoring with GPT-5 as judge

Significant gains in comparative, analytical, and predictive questions

Broader, more diverse retrieval with minimal losses in tail risk analysis

Abstract

Retrieval Augmented Generation (RAG) provides the necessary informational grounding to LLMs in the form of chunks retrieved from a vector database or through web search. RAG could also use knowledge graph triples as a means of providing factual information to an LLM. However, the retrieval is only based on representational similarity between a question and the contents. The performance of RAG depends on the numeric vector representations of the query and the chunks. To improve these representations, we propose Structured RAG (SRAG), which adds structured information to a query as well as the chunks in the form of topics, sentiments, query and chunk types (e.g., informational, quantitative), knowledge graph triples and semantic tags. Experiments indicate that this method significantly improves the retrieval process. Using GPT-5 as an LLM-as-a-judge, results show that the method improves the score given to answers in a question answering system by 30% (p-value = 2e-13) (with tighter bounds). The strongest improvement is in comparative, analytical and predictive questions. The results suggest that our method enables broader, more diverse, and episodic-style retrieval. Tail risk analysis shows that SRAG attains very large gains more often, with losses remaining minor in magnitude.