让检索到的段落更连贯，提升复杂问答准确率
CAGE: Coherence-Aware Graph Encoding for Retrieval-Augmented Generation
- 构建实体图模型，捕捉段落间的逻辑关系
- 在多跳问答中，精确匹配率显著提升
- 适合需要逻辑连贯性的长文本生成任务
传统检索增强生成系统独立评分每个段落，导致上下文虽个体相关却整体不连贯。我们提出一致性感知图编码（CAGE），通过四个维度建模段落间连贯性：域内相关性、抗噪声能力、信息关联性和事实一致性。该方法将检索到的段落转换为有向异构实体图，通过最小出度重加权强化事实锚点，利用关系图卷积网络编码结构模式，并融合段落间连贯性与查询相关性进行最终排序。在四个多跳推理基准上评估，CAGE在桥接型数据集上达到或超过monoT5的Recall@5，在下游任务中持续提升精确匹配率，表明即使检索召回率相当或更低，结构连贯的上下文也能产生更精准答案。
原文摘要 · Abstract (English)
Traditional Retrieval-Augmented Generation (RAG) systems score each passage independently against the query, assembling context sets that may be individually relevant yet collectively incoherent. We introduce Coherence-Aware Graph Encoding (CAGE), a reranking framework that models "between-chunk coherence" across four dimensions: Intra-Domain Relevance, Noise Resistance, Informational Bonding, and Factual Consistency. Our pipeline transforms retrieved passages into directed heterogeneous entity graphs, amplifies factual anchors via min-out-degree reweighting, encodes structural patterns through a Relational Graph Convolutional Network, and fuses inter-chunk coherence with query relevance for final ranking. Evaluated across four multi-hop benchmarks, CAGE matches or outperforms strong baselines including monoT5 in Recall@5 on bridge-dominated datasets and consistently improves downstream Exact Match, demonstrating that structurally coherent context yields more precise answers even when retrieval recall is comparable or lower.
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。