利用语义分块与混合检索策略改进长文档上下文检索
原标题：Improving Contextual Retrieval for Long Documents using Semantic Chunking and Hybrid Retrieval Strategies
AI 摘要
该论文提出一种混合 RAG 框架，将知识图谱的结构化推理与稠密向量检索的语义覆盖能力相结合，并采用基于 transformer 的摘要与语义分块多阶段流水线来提升长文档的上下文保留。系统在推理时优先进行图遍历，必要时回退到向量检索。在 MultiHop-RAG 数据集上的评估显示，该方法优于单独的向量或图基线，在复杂查询上取得更高的 Exact Match 分数和检索准确率。
正文节选
[type=editor, auid=1001,bioid=1] [auid=1002,bioid=2] Improving Contextual Retrieval for Long Documents using Semantic Chunking and Hybrid Retrieval Strategies Abstract The vast amount of information available in lengthy, distributed documents makes finding relevant answers difficult, especially when the answer requires connecting evidence from multiple sources. Standard Retrieval-Augmented Generation (RAG) systems often struggle with this type of ”multi-hop” reasoning and frequently face issues