通过逐项核查对话内容，有效减少大模型的幻觉问题。
Fine-Refine: Iterative Fine-grained Refinement for Mitigating Dialogue Hallucination
- 将对话拆解为最小事实单元，逐一验证
- 在两个数据集上使事实准确率提升7.63点
- 适合需要高可信度对话系统的研发者
当前大语言模型在对话系统中普遍存在幻觉问题，导致生成内容与事实不符，影响用户信任。现有修正方法多在整句层面操作，忽视单个回复中可能包含多个可验证或不可验证的事实。为此，我们提出Fine-Refine框架，将回复分解为原子单元，利用外部知识逐一验证，并通过困惑度评估流畅性，迭代修正细粒度错误。在HybriDialogue和OpendialKG数据集上评估事实性，以事实分数（fact score）和信息不足比例（Not Enough Information Proportion）为指标，实验表明Fine-Refine显著提升事实性，最高实现7.63点的事实分数提升，仅带来轻微对话质量损失。
原文摘要 · Abstract (English)
The tendency for hallucination in current large language models (LLMs) negatively impacts dialogue systems. Such hallucinations produce factually incorrect responses that may mislead users and undermine system trust. Existing refinement methods for dialogue systems typically operate at the response level, overlooking the fact that a single response may contain multiple verifiable or unverifiable facts. To address this gap, we propose Fine-Refine, a fine-grained refinement framework that decomposes responses into atomic units, verifies each unit using external knowledge, assesses fluency via perplexity, and iteratively corrects granular errors. We evaluate factuality across the HybriDialogue and OpendialKG datasets in terms of factual accuracy (fact score) and coverage (Not Enough Information Proportion), and experiments show that Fine-Refine substantially improves factuality, achieving up to a 7.63-point gain in dialogue fact score, with a small trade-off in dialogue quality.
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。