用句子级注意力优化检索，提升大模型问答的准确与效率
ParetoRAG: Leveraging Sentence-Context Attention for Robust and Efficient Retrieval-Augmented Generation
- 基于帕累托原则分解段落为句子，动态重加权核心内容
- 无需额外训练，在多个数据集上同时提升检索精度与生成质量
- 适合追求高效可靠知识增强的AI应用开发者
检索增强生成（RAG）系统虽能提升大语言模型（LLM）的外部知识利用能力，但仍面临检索效率低和模型无法过滤无关信息的问题。本文提出无监督框架ParetoRAG，通过句子级细化并遵循帕累托原则优化检索。将段落拆分为句子，动态调整关键内容权重，同时保持上下文连贯性，实现检索精度与生成质量双重提升，且无需额外训练或API资源。该框架已在多种数据集、大模型和检索器上得到实证验证。
原文摘要 · Abstract (English)
While Retrieval-Augmented Generation (RAG) systems enhance Large Language Models (LLMs) by incorporating external knowledge, they still face persistent challenges in retrieval inefficiency and the inability of LLMs to filter out irrelevant information. We present ParetoRAG, an unsupervised framework that optimizes RAG systems through sentence-level refinement guided by the Pareto principle. By decomposing paragraphs into sentences and dynamically re-weighting core content while preserving contextual coherence, ParetoRAG achieves dual improvements in both retrieval precision and generation quality without requiring additional training or API resources. This framework has been empirically validated across various datasets, LLMs, and retrievers.
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。