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.

翻译：检索增强生成（RAG）是一种前景广阔的范式，但其可信度仍是一个关键问题。一个主要的脆弱性出现在生成之前：模型往往难以平衡参数化（内部）知识与检索（外部）知识，尤其是在两种知识来源相互冲突或不可靠的情况下。为了全面分析这些场景，我们构建了包含36,266个问题、涵盖四种RAG设置的可信度响应数据集（TRD）。我们发现，现有方法仅能应对孤立场景——或优先考虑一种知识来源，或简单合并两者，或拒绝回答——但缺乏一个统一的框架来同时处理不同的现实世界条件。因此，我们提出了BRIDGE框架，该框架能动态确定大型语言模型（LLMs）的综合响应策略。BRIDGE利用一种名为软偏置的自适应加权机制来指导知识收集，随后通过最大软偏置决策树来评估知识并选择最优响应策略（信任内部/外部知识，或拒绝）。实验表明，BRIDGE在准确性上优于基线方法5-15%，同时在所有场景中保持均衡的性能。我们的工作为现实世界RAG应用中LLMs的可信响应提供了一个有效的解决方案。