KAMR: Grounding Generation via Knowledge-Aligned Multi-hop Retrieval
arXiv:2607.27136
The paper presents KAMR, a knowledge‑aligned multi‑hop retriever that first identifies anchor graph triplets strongly tied to a query and then locally expands to connected evidence, using a partially aligned dataset generated via LLM prompting and contrastive training to improve multi‑hop retrieval and downstream QA.
Abstract
Graph-based retrieval-augmented generation increasingly relies on multi-hop retrieval, where answering a query requires composing multiple connected knowledge-graph triplets. However, existing retrievers often rank triplets independently via global semantic matching. Moreover, many multi-hop benchmarks provide only final answers, which limits supervision for query--triplet alignment and causes structurally necessary but weakly aligned facts to be missed. To address these issues, we propose a knowledge-aligned multi-hop retriever, KAMR, which distinguishes anchor triplets that are strongly constrained by the query from connected triplets that are weakly aligned yet structurally linked to the anchors. To mitigate the lack of query-triplet alignment supervision, we build a partial alignment dataset by masking triplet elements and prompting an LLM to generate corresponding queries, and optimize two contrastive objectives for pair-level and element-level matching. At inference time, KAMR retrieves anchors globally and then expands locally to collect connected evidence. Across four benchmarks, three LLM backbones, and fourteen baselines, KAMR consistently improves multi-hop retrieval and downstream question answering performance.
Accepted by COLM'26