提出自适应树结构框架，提升长文档问答的推理与证据整合能力
A Tree-based RAG Framework for Evidence-Intensive QA via Adaptive Planning and Topology-Aware Evidence Gathering
- 基于动态规划扩展树结构，按问题依赖灵活推理
- 通过兄弟节点复用和子节点聚合，覆盖更广证据范围
- 支持批量生成答案，显著降低长流程问答的计算开销
现有结构化RAG方法虽利用树或图结构改进多跳问答，但在证据密集型问答中仍面临两大瓶颈：结构僵化导致推理扩展受限，拓扑无关的证据获取无法有效整合不同推理节点的信息。为此，我们提出APT-RAG框架，包含自适应规划与拓扑感知证据收集机制。自适应规划根据问题依赖和证据需求动态扩展推理结构；拓扑感知证据收集通过兄弟节点证据复用、直接检索及子节点证据聚合，提升证据覆盖度。此外，引入证据引导的批量答案生成策略，显著减少证据密集型问答中的生成开销。在多个证据密集型问答基准上的实验表明，APT-RAG优于现有结构化RAG方法。代码已开源：https://github.com/hyudsl/APT-RAG。
原文摘要 · Abstract (English)
Recent structured RAG methods leverage tree- or graph-based reasoning structures to improve multi-hop QA. However, they face key limitations in evidence-intensive QA, where answering a question requires synthesizing information scattered across dozens or even hundreds of documents: structural rigidity, which limits adaptive reasoning expansion, and topology-ignorant evidence gathering, which prevents effective integration of evidence across different reasoning nodes. To address these issues, we propose APT-RAG, an Adaptive Planning and Topology-aware evidence gathering RAG framework. Adaptive planning dynamically expands the reasoning structure based on question dependencies and evidence requirements, while topology-aware evidence gathering improves evidence coverage through sibling evidence reuse, direct retrieval, and evidence aggregation from child nodes. We further introduce evidence-guided batched answer generation to reduce significant generation overhead in evidence-intensive QA. In the experiments on evidence-intensive QA benchmarks, APT-RAG outperforms existing structured RAG methods. Our code is available at https://github.com/hyudsl/APT-RAG.
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。