提出高效多跳检索增强生成框架，减少重复计算与无效推理。
Think Straight, Stop Smart: Structured Reasoning for Efficient Multi-Hop RAG
- 用模板化推理缓存共用前缀，锚定子问题到主问题
- 基于检索器的确定性终止机制，避免随机停顿
- 适合资源受限场景，如设备端推理
多跳检索增强生成（RAG）是复杂推理的有前景策略，但现有迭代提示方法效率低下。它们常在每一步重复生成可预测的词元序列，并依赖随机停止机制，导致词元消耗过多且终止不稳定。我们提出TSSS（Think Straight, Stop Smart）——一种面向高效的结构化多跳RAG框架。TSSS引入两项设计：(i) 模板化推理，通过缓存重复前缀并将子问题锚定至主问题，降低生成成本并提升推理稳定性；(ii) 基于检索器的终止器，一旦新增子问题出现重复即确定性终止。该结构分离推理与终止控制，实现更快推理和更可靠答案。在HotpotQA、2WikiMultiHop和MuSiQue数据集上，TSSS达到当前最优准确率，在RAG-CoT方法中兼具竞争力效率，凸显其在资源受限场景（如设备端推理）中的有效性。
原文摘要 · Abstract (English)
Multi-hop retrieval-augmented generation (RAG) is a promising strategy for complex reasoning, yet existing iterative prompting approaches remain inefficient. They often regenerate predictable token sequences at every step and rely on stochastic stopping, leading to excessive token usage and unstable termination. We propose TSSS (Think Straight, Stop Smart), a structured multi-hop RAG framework designed for efficiency. TSSS introduces (i) a template-based reasoning that caches recurring prefixes and anchors sub-queries to the main question, reducing token generation cost while promoting stable reasoning, and (ii) a retriever-based terminator, which deterministically halts reasoning once additional sub-queries collapse into repetition. This separation of structured reasoning and termination control enables both faster inference and more reliable answers. On HotpotQA, 2WikiMultiHop, and MuSiQue, TSSS achieves state-of-the-art accuracy and competitive efficiency among RAG-CoT approaches, highlighting its effectiveness in efficiency-constrained scenarios such as on-device inference.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。