让大模型在检索生成时能看清知识冲突并自主判断，提升回答准确性。
Seeing through the Conflict: Transparent Knowledge Conflict Handling in Retrieval-Augmented Generation
- 用双对比编码器分离语义匹配与事实一致性，精准识别矛盾
- 通过自答能力评估内部记忆可信度，减少幻觉和错误依赖
- 轻量级软提示融合三类信号，效果媲美大模型且不增加计算负担
配备检索的大语言模型（如RAG）本应结合自身参数化知识与外部证据，但实际中常产生幻觉、过度信任噪声片段或忽略关键上下文。本文提出TCR（透明冲突解析）框架，使决策过程可观察、可控制。TCR通过双对比编码器解耦语义匹配与事实一致性，估算自答能力以衡量对内存储存的信心，并将三个标量信号通过基于信噪比加权的轻量软提示输入生成器。在七个基准测试中，TCR将冲突检测F1提升5-18%，知识缺口恢复率提高21.4个百分点，误导性上下文覆盖降低29.3个百分点，仅增加0.3%参数量。信号与人类判断高度一致，并揭示了决策的时间演化模式。
原文摘要 · Abstract (English)
Large language models (LLMs) equipped with retrieval--the Retrieval-Augmented Generation (RAG) paradigm--should combine their parametric knowledge with external evidence, yet in practice they often hallucinate, over-trust noisy snippets, or ignore vital context. We introduce TCR (Transparent Conflict Resolution), a plug-and-play framework that makes this decision process observable and controllable. TCR (i) disentangles semantic match and factual consistency via dual contrastive encoders, (ii) estimates self-answerability to gauge confidence in internal memory, and (iii) feeds the three scalar signals to the generator through a lightweight soft-prompt with SNR-based weighting. Across seven benchmarks TCR improves conflict detection (+5-18 F1), raises knowledge-gap recovery by +21.4 pp and cuts misleading-context overrides by -29.3 pp, while adding only 0.3% parameters. The signals align with human judgements and expose temporal decision patterns.
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。