Democratizing GraphRAG: Linear, CPU-Only Graph Retrieval for Multi-Hop QA
This work democratizes GraphRAG by enabling efficient multi-hop retrieval without token costs or GPU requirements, though it is incremental as it builds on existing graph retrieval methods.
The paper tackled the problem of expensive LLM-based graph construction and GPU-heavy inference in GraphRAG systems for multi-hop QA by introducing SPRIG, a CPU-only, linear-time pipeline that uses NER-driven co-occurrence graphs and Personalized PageRank, achieving a 28% improvement in recall with negligible changes in Recall@10.
GraphRAG systems improve multi-hop retrieval by modeling structure, but many approaches rely on expensive LLM-based graph construction and GPU-heavy inference. We present SPRIG (Seeded Propagation for Retrieval In Graphs), a CPU-only, linear-time, token-free GraphRAG pipeline that replaces LLM graph building with lightweight NER-driven co-occurrence graphs and uses Personalized PageRank (PPR) for 28% with negligible Recall@10 changes. The results characterize when CPU-friendly graph retrieval helps multi-hop recall and when strong lexical hybrids (RRF) are sufficient, outlining a realistic path to democratizing GraphRAG without token costs or GPU requirements.