通过融合模型不确定性，减少大模型幻觉生成。
Uncertainty-Aware Fusion: An Ensemble Framework for Mitigating Hallucinations in Large Language Models
- 基于多模型准确率与自评估能力，动态融合输出。
- 在多个基准上提升8%事实准确性，逼近GPT-4表现。
- 适合需要高可信度文本生成的场景，如医疗、法律。
大型语言模型（LLMs）常产生非事实性输出，影响用户信任。传统方法如表示编辑和对比解码需额外训练数据且实现复杂。基于集成的方法虽利用多模型“群体智慧”，却忽略单个模型输出的不确定性。近期研究发现，不确定性估计可使模型自我评估幻觉可能性。本文聚焦事实型问答任务，观察到不同模型在不同场景下准确率与自评估能力差异显著。据此提出不确定性感知融合（UAF）框架，通过结合各模型的准确率与自评估能力，策略性融合输出以降低幻觉。在多个公开基准数据集上的实证结果表明，UAF在事实准确性上优于现有最优幻觉缓解方法8%，同时缩小或超越与GPT-4的性能差距。
原文摘要 · Abstract (English)
Large Language Models (LLMs) are known to hallucinate and generate non-factual outputs which can undermine user trust. Traditional methods to directly mitigate hallucinations, such as representation editing and contrastive decoding, often require additional training data and involve high implementation complexity. While ensemble-based approaches harness multiple LLMs to tap into the "wisdom of crowds", these methods overlook uncertainties in individual model responses. Recent studies reveal that uncertainty estimation can enable LLMs to self-assess the likelihood of generating hallucinations. In this work, we focus on factoid question answering (QA) and observe that LLMs accuracy and self-assessment capabilities vary widely with different models excelling in different scenarios. Leveraging this insight, we propose Uncertainty-Aware Fusion (UAF), an ensemble framework to reduces hallucinations by strategically combining multiple LLM based on their accuracy and self-assessment abilities. Empirical results on several public benchmark datasets show that UAF outperforms state-of-the-art hallucination mitigation methods by $8\%$ in factual accuracy, while either narrowing or surpassing the performance gap with GPT-4.
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。