Search papers, labs, and topics across Lattice.
LineageRAG enhances GraphRAG by explicitly constructing evidence lineages that connect evidence discovery to source grounding for multi-hop questions. This method improves the retrieval process by associating evidence demands with corresponding source spans, ensuring that the retrieved evidence is both relevant and verifiable. Experiments on benchmark datasets like HotpotQA and MuSiQue demonstrate significant performance gains, with average improvements of 3.51 in R@5, 5.96 in EM, and 5.22 in F1 over existing GraphRAG approaches.
Evidence lineages in LineageRAG not only improve retrieval accuracy but also ensure that every piece of evidence is grounded in verifiable source text.
Graph-based Retrieval-Augmented Generation (GraphRAG) retrieves evidence for multi-hop questions over structured cor- pus graphs. Existing GraphRAG methods leave the connection between evidence discovery and source grounding implicit. We propose LineageRAG, which constructs one evidence lin- eage for each query-derived evidence demand and completes it with a verbatim source span when the selected evidence supports that demand. LineageRAG first initializes the evi- dence demands. It then expands each lineage through demand- conditioned retrieval over the corpus graph while retaining the demand associated with every candidate. Lineage completion uses this provenance to select complementary passages and grounds supported demands in verbatim source text. Experi- ments on HotpotQA, 2WikiMultiHopQA, and MuSiQue show that LineageRAG improves R@5, EM, and F1 by 3.51, 5.96, and 5.22 points on average over leading GraphRAG baselines.
