Hi-Q：面向多跳问答的层级证据引导查询改写

摘要

多跳问答（QA）的一个核心瓶颈在于，问题的表达粒度通常与语料证据的可检索粒度不一致。现有方法通过在语料上施加固定图结构、迭代地改写查询，或执行一个在语料上生成的程序来应对这种不匹配；但这些策略并未明确判断一个查询单元何时已被证据支持、何时应当被进一步细化。我们将这一瓶颈形式化为“可检索粒度发现”，并引入Hi-Q——一种以证据为条件的分层查询细化框架。在每个查询节点上，一个解析算子会测试检索到的证据是否支持当前查询单元；已解析的节点终止，未解析的节点则由一个保持依赖关系的二元算子进行扩展，并通过语义覆盖验证器检验。因此，Hi-Q生成的查询树，其拓扑结构由语料支持信号决定，而非由固定分解模板或预构建图决定。我们在三个多跳问答基准上评估了Hi-Q，主要采用全语料检索设置；在该设置中，答案所依赖的证据必须从开放域干扰项中定位，而非取自小型标注池。在此设置下，Hi-Q在三个基准上的平均EM为52.3、F1为64.0；在相同平均指标上，比迭代检索基线IRCoT高出15.1 EM / 18.2 F1；在MuSiQue-full上，比基于图的RAG基线PropRAG高出11.5 EM / 12.0 F1，且无需构建全语料图。在先前工作所用的受限支持/干扰项设置中，Hi-Q同样取得了最佳准确率，平均EM为57.9、F1为69.3，比PropRAG高出5.6 EM / 3.9 F1，比IRCoT高出13.7 EM / 15.8 F1。项目页面见 https://hi-q-project.github.io/。

English

A central bottleneck in multi-hop Question Answering (QA) is that the granularity at which a question is expressed often differs from the granularity at which corpus evidence is retrievable. Existing methods address this mismatch by imposing fixed graph structures over the corpus, by iteratively reformulating the query, or by executing a generated program over it, but these strategies do not explicitly decide when a query unit is already supported by evidence and when it should be refined. We formulate this bottleneck as retrievable granularity discovery and introduce Hi-Q, an evidence-conditioned framework for hierarchical query refinement. At each query node, a resolution operator tests whether retrieved evidence supports the current query unit; resolved nodes terminate, while unresolved nodes are expanded by a dependency-preserving binary operator and checked by a semantic coverage verifier. Hi-Q therefore grows a query tree whose topology is determined by corpus support signals rather than by a fixed decomposition template or a pre-built graph. We evaluate Hi-Q on three multi-hop QA benchmarks, primarily under full-corpus retrieval, where dependent evidence must be located among open-domain distractors rather than within a small annotated pool. In this setting Hi-Q reaches 52.3 EM and 64.0 F1 averaged over the three benchmarks, ahead of the iterative retrieval baseline IRCoT by 15.1 EM / 18.2 F1 on that same average, and ahead of the graph-based RAG baseline PropRAG by 11.5 EM / 12.0 F1 on MuSiQue-full, without corpus-wide graph construction. In the restricted supporting/distractor setting used by prior work, Hi-Q likewise attains the best accuracy, with 57.9 EM and 69.3 F1 on average, ahead of PropRAG by 5.6 EM / 3.9 F1 and IRCoT by 13.7 EM / 15.8 F1. The project page is available at https://hi-q-project.github.io/.