SlimRAG: Retrieval without Graphs via Entity-Aware Context Selection

TL;DR

SlimRAG introduces a lightweight, entity-aware retrieval framework for language models that improves accuracy and efficiency by avoiding complex graph structures and focusing on relevant entities and context.

Contribution

It proposes a novel, structure-free retrieval method using entity embeddings and a new metric RITU, outperforming graph-based methods in accuracy and efficiency.

Findings

Outperforms graph-based baselines in accuracy across QA benchmarks.

Reduces index size and retrieval overhead significantly.

Achieves higher RITU scores indicating more compact and relevant retrievals.

Abstract

Retrieval-Augmented Generation (RAG) enhances language models by incorporating external knowledge at inference time. However, graph-based RAG systems often suffer from structural overhead and imprecise retrieval: they require costly pipelines for entity linking and relation extraction, yet frequently return subgraphs filled with loosely related or tangential content. This stems from a fundamental flaw -- semantic similarity does not imply semantic relevance. We introduce SlimRAG, a lightweight framework for retrieval without graphs. SlimRAG replaces structure-heavy components with a simple yet effective entity-aware mechanism. At indexing time, it constructs a compact entity-to-chunk table based on semantic embeddings. At query time, it identifies salient entities, retrieves and scores associated chunks, and assembles a concise, contextually relevant input -- without graph traversal or edge construction. To quantify retrieval efficiency, we propose Relative Index Token Utilization (RITU), a metric measuring the compactness of retrieved content. Experiments across multiple QA benchmarks show that SlimRAG outperforms strong flat and graph-based baselines in accuracy while reducing index size and RITU (e.g., 16.31 vs. 56+), highlighting the value of structure-free, entity-centric context selection. The code will be released soon. https://github.com/continue-ai-company/SlimRAG