发表机构
LinYi University; Harbin Institute of Technology（临沂大学; 哈尔滨工业大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
针对检索增强生成中知识冲突导致的不可靠响应，提出TRACE微调框架，利用多智能体辩论轨迹和答案完整性正则化，增强知识源选择与冲突稳健性，减少不完整答案。
AI 中文摘要
检索增强生成（RAG）通过引入外部上下文来缓解大语言模型中的知识过时和事实幻觉问题。然而，当检索到的知识与模型的内部参数知识发生冲突时，模型可能盲目遵循误导性上下文，或错误地依赖参数知识，导致响应不可靠。为解决这一问题，本文提出TRACE（辩论轨迹与答案完整性正则化微调），一种面向知识冲突下RAG的稳健微调框架。首先，我们提出一种利用多智能体辩论轨迹提取正确候选、错误候选和答案偏移模式的微调方法，为可靠的知识源选择提供细粒度监督。此外，我们设计了答案完整性正则化机制，通过答案尾部标记强化和过早终止抑制来缓解空响应、过短响应和提前终止的响应。微调目标结合了正确答案监督、错误候选抑制、答案尾部标记强化和过早终止抑制，使模型能够使用可靠的外部上下文，抵抗误导性或无关的检索内容，并在检索证据不可靠时回退到参数知识。跨多个知识冲突场景和数据集的实验表明，TRACE提高了对误导性检索知识的稳健性，并减少了不完整答案。这些结果证明，多智能体辩论轨迹和答案完整性正则化共同增强了RAG模型中知识源选择、冲突稳健性和答案质量。我们的代码可在以下网址获取：此https URL。
英文摘要
Retrieval-Augmented Generation (RAG) mitigates knowledge obsolescence and factual hallucination in large language models by introducing external context. However, when retrieved knowledge conflicts with the model's internal parametric knowledge, the model may either blindly follow misleading context or incorrectly rely on parametric knowledge, leading to unreliable responses. To address this issue, this paper proposes TRACE (Debate-TRace and Answer-Completeness rEgularized fine-tuning), a robust fine-tuning framework for RAG under knowledge conflicts. First, we propose a fine-tuning method that leverages multi-agent debate traces to extract correct candidates, incorrect candidates, and answer-shift patterns, providing fine-grained supervision for reliable knowledge-source selection. In addition, we design an answer completeness regularization mechanism to alleviate empty, overly short, and prematurely terminated responses via answer-tail token reinforcement and premature termination suppression. The fine-tuning objective combines correct-answer supervision, incorrect-candidate suppression, answer-tail token reinforcement, and premature termination suppression, enabling the model to use reliable external context, resist misleading or irrelevant retrieved content, and fall back to parametric knowledge when retrieved evidence is unreliable. Experiments across multiple knowledge-conflict scenarios and datasets show that TRACE improves robustness against misleading retrieved knowledge and reduces incomplete answers. These results demonstrate that multi-agent debate traces and answer completeness regularization jointly enhance knowledge-source selection, conflict robustness, and answer quality in RAG models. Our code is available at https://github.com/PHD-lanyu/TRACE.
CommentsAccepted by ICDM 2026
