GraphER: An Efficient Graph-Based Enrichment and Reranking Method for Retrieval-Augmented Generation

arXiv cs.LG / 3/27/2026

📰 NewsIdeas & Deep AnalysisModels & Research

共有:

Key Points

The paper argues that semantic search inside RAG systems can fail when evidence is scattered across multiple sources, and that agentic retrieval can be inefficient because it expands search iteratively without fully exploiting data structure.
It proposes GraphER, which enriches data offline during indexing and then performs graph-based reranking at query time to capture multiple proximities beyond raw semantic similarity.
GraphER avoids the need for maintaining a knowledge graph and is designed to work directly with standard vector stores commonly used in production.
The method is retriever-agnostic and is reported to add negligible latency overhead, with experiments showing improved performance across several retrieval benchmarks.

Abstract

Semantic search in retrieval-augmented generation (RAG) systems is often insufficient for complex information needs, particularly when relevant evidence is scattered across multiple sources. Prior approaches to this problem include agentic retrieval strategies, which expand the semantic search space by generating additional queries. However, these methods do not fully leverage the organizational structure of the data and instead rely on iterative exploration, which can lead to inefficient retrieval. Another class of approaches employs knowledge graphs to model non-semantic relationships through graph edges. Although effective in capturing richer proximities, such methods incur significant maintenance costs and are often incompatible with the vector stores used in most production systems. To address these limitations, we propose GraphER, a graph-based enrichment and reranking method that captures multiple forms of proximity beyond semantic similarity. GraphER independently enriches data objects during offline indexing and performs graph-based reranking over candidate objects at query time. This design does not require a knowledge graph, allowing GraphER to integrate seamlessly with standard vector stores. In addition, GraphER is retriever-agnostic and introduces negligible latency overhead. Experiments on multiple retrieval benchmarks demonstrate the effectiveness of the proposed approach.