旅游问答中用查询扩展重构候选段落图，提升大模型检索准确性。
QCG-Rerank: Chunks Graph Rerank with Query Expansion in Retrieval-Augmented LLMs for Tourism Domain
- 通过提取关键信息扩展原始查询，增强语义匹配
- 构建段落图并迭代计算转移概率，筛选最高分段落
- 在多个数据集上优于传统RAG，适合短查询场景
检索增强生成（RAG）通过引入信息检索缓解大语言模型的幻觉问题。但在旅游领域，查询通常简短而数据库内容多样，现有RAG检索后可能包含大量无关或矛盾信息。为此，我们提出QCG-Rerank模型：先进行初始检索获取候选段落，再提取关键信息扩展原查询；利用扩展查询与候选段落计算相似度得分作为初始转移概率，构建段落图；随后基于初始估计迭代计算转移概率直至收敛；最终选取得分最高的段落输入LLM生成回答。我们在Cultour、IIRC、StrategyQA、HotpotQA、SQuAD和MuSiQue数据集上评估，实验结果证明该方法有效且优越。
原文摘要 · Abstract (English)
Retrieval-Augmented Generation (RAG) mitigates the issue of hallucination in Large Language Models (LLMs) by integrating information retrieval techniques. However, in the tourism domain, since the query is usually brief and the content in the database is diverse, existing RAG may contain a significant amount of irrelevant or contradictory information contents after retrieval. To address this challenge, we propose the QCG-Rerank model. This model first performs an initial retrieval to obtain candidate chunks and then enhances semantics by extracting critical information to expand the original query. Next, we utilize the expanded query and candidate chunks to calculate similarity scores as the initial transition probability and construct the chunks graph. Subsequently, We iteratively compute the transition probabilities based on an initial estimate until convergence. The chunks with the highest score are selected and input into the LLMs to generate responses. We evaluate the model on Cultour, IIRC, StrategyQA, HotpotQA, SQuAD, and MuSiQue datasets. The experimental results demonstrate the effectiveness and superiority of the QCG-Rerank method.
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。