arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2601.00536cs.CL
收藏

多跳问答中的检索-推理过程:一种四维设计框架和实证趋势

Retrieval--Reasoning Processes for Multi-hop Question Answering: A Four-Axis Design Framework and Empirical Trends

School of Computing and Information, University of Pittsburgh(计算与信息学院,匹兹堡大学);Google Cloud AI Research(谷歌云人工智能研究)

Yuelyu Ji, Zhuochun Li, Rui Meng, Daqing He

AI总结:

本文提出四维设计框架,分析多跳问答系统中检索-推理过程的执行计划、索引结构、控制策略和停止标准,揭示有效性、效率与证据忠实度的权衡,并指出结构感知规划、可转移控制策略和鲁棒停止是未来挑战。

摘要:

多跳问答需要系统迭代检索证据并在多个步骤中推理。尽管最近的RAG和代理方法报告了强劲的结果,但底层的检索-推理过程往往被忽略,使得在不同模型家族之间比较过程选择变得困难。本文将执行过程作为分析单位,引入一个四维框架,涵盖(A)总体执行计划,(B)索引结构,(C)下一步控制(策略和触发器),以及(D)停止/继续标准。使用此框架,我们映射了代表性的多跳问答系统,并综合了在标准基准(如HotpotQA、2WikiMultiHopQA、MuSiQue)上报告的消融分析和趋势,突显了有效性、效率和证据忠实度之间的反复权衡。最后,我们总结了检索-推理代理的开放挑战,包括结构感知的规划、可转移的控制策略以及在分布偏移下的鲁棒停止。

英文摘要:

Multi-hop question answering (QA) requires systems to iteratively retrieve evidence and reason across multiple hops. While recent RAG and agentic methods report strong results, the underlying retrieval--reasoning \emph{process} is often left implicit, making procedural choices hard to compare across model families. This survey takes the execution procedure as the unit of analysis and introduces a four-axis framework covering (A) overall execution plan, (B) index structure, (C) next-step control (strategies and triggers), and (D) stop/continue criteria. Using this schema, we map representative multi-hop QA systems and synthesize reported ablations and tendencies on standard benchmarks (e.g., HotpotQA, 2WikiMultiHopQA, MuSiQue), highlighting recurring trade-offs among effectiveness, efficiency, and evidence faithfulness. We conclude with open challenges for retrieval--reasoning agents, including structure-aware planning, transferable control policies, and robust stopping under distribution shift.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑