解决RAG中内外知识冲突时模型不靠谱的问题
After Retrieval, Before Generation: Enhancing the Trustworthiness of Large Language Models in Retrieval-Augmented Generation
- 用动态权重机制融合内外知识，避免盲目信任任一来源
- 在3.6万题数据集上提升准确率5-15%，全场景表现均衡
- 适合追求真实可靠生成的AI应用开发者
检索增强生成（RAG）虽有潜力，但可信度仍是关键挑战。主要风险出现在生成前：模型常无法平衡参数化（内部）与检索到（外部）的知识，尤其当两者冲突或不可靠时。为此，我们构建了涵盖四种RAG场景的可信响应数据集（TRD），包含36,266个问题。分析发现，现有方法仅针对单一场景——优先内部/外部知识、简单合并或拒绝回答——缺乏统一框架应对复杂现实条件。因此，我们提出BRIDGE框架，通过自适应加权机制（soft bias）引导知识收集，并采用最大软偏倚决策树评估知识，选择最优响应策略（信任内部、信任外部或拒绝）。实验显示，BRIDGE在准确率上优于基线5-15%，且在所有场景中保持平衡表现。本工作为真实RAG应用中大模型的可信响应提供了有效方案。
原文摘要 · Abstract (English)
Retrieval-augmented generation (RAG) is a promising paradigm, yet its trustworthiness remains a critical concern. A major vulnerability arises prior to generation: models often fail to balance parametric (internal) and retrieved (external) knowledge, particularly when the two sources conflict or are unreliable. To analyze these scenarios comprehensively, we construct the Trustworthiness Response Dataset (TRD) with 36,266 questions spanning four RAG settings. We reveal that existing approaches address isolated scenarios-prioritizing one knowledge source, naively merging both, or refusing answers-but lack a unified framework to handle different real-world conditions simultaneously. Therefore, we propose the BRIDGE framework, which dynamically determines a comprehensive response strategy of large language models (LLMs). BRIDGE leverages an adaptive weighting mechanism named soft bias to guide knowledge collection, followed by a Maximum Soft-bias Decision Tree to evaluate knowledge and select optimal response strategies (trust internal/external knowledge, or refuse). Experiments show BRIDGE outperforms baselines by 5-15% in accuracy while maintaining balanced performance across all scenarios. Our work provides an effective solution for LLMs' trustworthy responses in real-world RAG applications.
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。