Search papers, labs, and topics across Lattice.
This paper introduces EviReform, a novel approach for multi-hop graph retrieval that enhances evidence-guided query reformulation by separating the processes of revising retrieval requests and aggregating evidence. By leveraging retrieved passages to formulate residual queries, EviReform effectively normalizes and combines retrieval signals, allowing for improved access to complementary evidence. The method demonstrates significant performance gains on benchmark datasets, achieving up to 5.59 Recall@5 points and 4.50 F1 points over the strongest baseline, underscoring the importance of observed evidence in guiding retrieval processes.
EviReform reveals that leveraging evidence from retrieved passages can significantly boost multi-hop retrieval performance, achieving up to 5.59 points in Recall@5.
Multi-hop retrieval must recover passages that provide sufficient evidence together. An initial passage often resolves an entity or relation implicit in the question, making the missing evidence easier to describe only after retrieval begins. Graph retrieval improves access to related evidence through stored corpus structure, but its retrieval signal is commonly derived from the original question. Complementary evidence must then be reached through stored relations even when an observed passage provides a more direct semantic cue. We introduce EviReform, which separates revising the retrieval request from aggregating evidence in the graph. Retrieved source passages formulate residual queries for the unresolved information need. The original and residual retrieval signals are normalized separately, combined, and propagated between propositions that share entities. On 2WikiMultiHopQA, HotpotQA, and MuSiQue, EviReform exceeds the strongest baseline by up to 5.59 Recall@5 points and 4.50 F1 points. These results show that observed evidence can guide graph retrieval toward the part of a supporting chain left underspecified by the original question. Code is available at https://github.com/XrazyMee/EviReform.
