通过动态构建知识图谱提升大模型推理能力
RAS: Retrieval-And-Structuring for Knowledge-Intensive LLM Generation
- 根据问题动态构建专属知识图谱，分步组织检索信息
- 在7个基准上显著超越基线，提升最高达8.7%
- 适合需要复杂推理的问答与知识密集型任务
大语言模型在知识密集型任务中表现优异，但在多步推理中常因检索内容无结构而受阻。尽管检索增强生成（RAG）提供外部信息，但片段间缺乏显式组织，导致推理路径脆弱。近期可解释性研究也表明结构化中间推理至关重要。本文提出检索与结构化框架（RAS），通过迭代检索与增量图构建，动态创建针对具体问题的知识图谱。RAS在七个知识密集型基准上持续优于强基线，使用专有模型时性能提升最高达8.7%，开源模型提升7.0%。结果表明，动态、问题相关的知识结构化是提升语言模型推理准确性和鲁棒性的有效路径。
原文摘要 · Abstract (English)
Large language models (LLMs) have achieved impressive performance on knowledge-intensive tasks, yet they often struggle with multi-step reasoning due to the unstructured nature of retrieved context. While retrieval-augmented generation (RAG) methods provide external information, the lack of explicit organization among retrieved passages limits their effectiveness, leading to brittle reasoning pathways. Recent interpretability studies highlighting the importance of structured intermediate reasoning further align with this perspective. We propose Retrieval-And-Structuring (RAS), a framework that dynamically constructs question-specific knowledge graphs through iterative retrieval and structured knowledge building. RAS interleaves targeted retrieval planning with incremental graph construction, enabling models to assemble and reason over evolving knowledge structures tailored to each query. On seven knowledge-intensive benchmarks, RAS consistently outperforms strong baselines, achieving up to 8.7\% and 7.0\% gains with proprietary and open-source LLMs, respectively. Our results demonstrate that dynamic, question-specific knowledge structuring offers a robust path to improving reasoning accuracy and robustness in language model generation.
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。