让检索文档更贴合问题，提升大模型生成的准确性。
Align Documents to Questions: Question-Oriented Document Rewriting for Retrieval-Augmented Generation
- 用风格引导重写技术，让文档更匹配问题语调。
- 在多个RAG系统中提升准确率，最高达8%相对改进。
- 可直接插入现有流程，几乎不增加延迟，适合部署使用。
检索增强生成（RAG）通过引入检索到的文档或生成上下文来提升大语言模型（LLM）的事实性。然而，当面对混合上下文时，LLM常表现出风格偏差，更倾向于流畅但虚构的生成内容，而非事实准确但结构松散的检索证据。这表明检索信息的效用受限于其呈现方式。为此，我们提出QREAM，一种风格可控的重写器，能在保持事实的前提下，将检索文档对齐至问题导向风格，更利于LLM利用。该框架包含两阶段：(1) QREAM-ICL，利用风格种子引导迭代重写探索；(2) QREAM-FT，一个从去噪ICL输出中蒸馏出的轻量级学生模型。QREAM-FT采用双准则拒绝采样，基于答案正确性和事实一致性进行过滤，确保高质量监督。QREAM可无缝集成至现有RAG流水线，作为即插即用模块。实验表明，QREAM持续提升先进RAG系统性能，最高实现8%相对改进，且延迟开销极小，有效平衡了问题相关性与事实准确性。
原文摘要 · Abstract (English)
Retrieval-Augmented Generation (RAG) enhances the factuality of Large Language Models (LLMs) by incorporating retrieved documents and/or generated context. However, LLMs often exhibit a stylistic bias when presented with mixed contexts, favoring fluent but hallucinated generated content over factually grounded yet disorganized retrieved evidence. This phenomenon reveals that the utility of retrieved information is bottlenecked by its presentation. To bridge this gap, we propose QREAM, a style-controlled rewriter that aligns retrieved documents with a question-oriented style while preserving facts, better for LLM readers to utilize. Our framework consists of two stages: (1) QREAM-ICL, which uses stylistic seeds to guide iterative rewriting exploration; and (2) QREAM-FT, a lightweight student model distilled from denoised ICL outputs. QREAM-FT employs dual-criteria rejection sampling, filtering based on answer correctness and factual consistency to ensure high-quality supervision. QREAM seamlessly integrates into existing RAG pipelines as a plug-and-play module. Experiments demonstrate that QREAM consistently enhances advanced RAG pipelines, yielding up to 8% relative improvement with negligible latency overhead, effectively balancing question relevance with factual grounding.
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。