用软标签蒸馏降低大模型幻觉，提升事实准确性。
Smoothing Out Hallucinations: Mitigating LLM Hallucination with Smoothed Knowledge Distillation
- 用教师模型生成平滑软标签，替代硬标签训练学生模型。
- 在摘要任务上幻觉率显著下降，通用NLP性能保持不变。
- 适合关注模型可靠性与事实一致性的研究者使用。
大型语言模型（LLMs）常出现幻觉，生成与事实不符或缺乏依据的内容，限制其在高风险场景中的应用。导致幻觉的关键因素之一是训练中使用硬标签，这带来确定性监督、诱发过度自信，并忽略自然语言中的不确定性。为此，我们提出通过知识蒸馏（KD）缓解幻觉：教师模型向学生模型提供平滑的软标签，降低过信心，增强事实依据。我们在指令数据上进行监督微调时应用该方法，评估了不同模型家族的效果。在摘要基准上的实验表明，相比标准微调，该方法有效减少幻觉，同时保持通用NLP任务性能。结果表明，知识蒸馏是缓解大模型幻觉、提升可靠性的有效途径。
原文摘要 · Abstract (English)
Large language models (LLMs) often suffer from hallucination, generating factually incorrect or ungrounded content, which limits their reliability in high-stakes applications. A key factor contributing to hallucination is the use of hard labels during training, which enforce deterministic supervision, encourage overconfidence, and disregard the uncertainty inherent in natural language. To address this, we propose mitigating hallucination through knowledge distillation (KD), where a teacher model provides smoothed soft labels to a student model, reducing overconfidence and improving factual grounding. We apply KD during supervised finetuning on instructional data, evaluating its effectiveness across LLMs from different families. Experimental results on summarization benchmarks demonstrate that KD reduces hallucination compared to standard finetuning while preserving performance on general NLP tasks. These findings highlight KD as a promising approach for mitigating hallucination in LLMs and improving model reliability.
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。