两篇新研究论文探讨了理解和量化大型语言模型（LLMs）中不确定性的方法。第一篇论文提出了一个框架，将LLM的不确定性分解为输入歧义、知识差距和解码随机性，从而深入了解模型的可靠性和幻觉检测。第二篇论文介绍了一种使用证据知识蒸馏的方法，无需传统采样方法的高计算成本即可实现高效的不确定性估计，仅需一次前向传播即可达到可比的性能。 AI
影响 这些论文在理解LLM可靠性和检测幻觉方面取得了进展，有望带来更值得信赖的AI系统。
排序理由 两篇在arXiv上发表的学术论文，讨论了LLM中不确定性量化方法。
AI 生成摘要 · Google Gemini · 来自 2 个来源。 我们如何撰写摘要 →
arXiv:2603.24967v2 Announce Type: replace Abstract: Understanding why a large language model (LLM) is uncertain about the response is important for their reliable deployment. Current approaches, which either provide a single uncertainty score or rely on the classical aleatoric-ep…
arXiv:2507.18366v2 Announce Type: replace-cross Abstract: Accurate uncertainty quantification remains a key challenge for standard LLMs, prompting the adoption of Bayesian and ensemble-based methods. However, such methods typically necessitate computationally expensive sampling, …