KNIT: Recovering the Passage That Top-k Truncation Cuts

Retrieval-augmented generation (RAG) passes the top $k$ passages of a retrieval pipeline to the LLM that answers. For a multi-hop question, the evidence is complete only when the top $k$ holds every passage of the question's evidence chain. But top-$k$ selection scores each passage on its own, by how well it matches the question. The passage that completes a chain is often linked to the others only through a bridge entity the question never names, so it ranks low and falls outside the top $k$. We introduce KNIT, a training-free layer that selects the top $k$ as a set. KNIT keeps the pipeline's relevance scores and adds links between passages that share an entity, read from a graph built once per corpus. A graph stage selects $2k$ linked passages that together cover the question, and one LLM call keeps $k$ of them. We apply one configuration, fixed on MuSiQue, to five published retrieval pipelines on five corpora. We measure chain completion: the share of questions whose top 5 holds the whole chain. KNIT raises it on all 25 pipeline–corpus pairs, significantly on 23. On the three Wikipedia corpora, the average gain is 10.6 to 13.7 points. KNIT improves even PropRAG, the strongest pipeline on Wikipedia, which already reads such a graph during retrieval. KNIT's gains carry over to question answering on Wikipedia: a 72B reader answers significantly more questions from KNIT's passages on 13 of the 15 Wikipedia pairs. On MuSiQue, our analysis shows that the graph stage's gain comes from the entity links rather than from diversity.