Uncertainty decomposition refers to the task of decomposing the total uncertainty of a model into data (aleatoric) uncertainty, resulting from the inherent complexity or ambiguity of the data, and model (epistemic) uncertainty, resulting from the lack of knowledge in the model. Performing uncertainty decomposition for large language models (LLMs) is an important step toward improving the reliability, trustworthiness, and interpretability of LLMs, but this research task is very challenging and remains unresolved. The existing canonical method, Bayesian Neural Network (BNN), cannot be applied to LLMs, because BNN requires training and ensembling multiple variants of models, which is infeasible or prohibitively expensive for LLMs. In this paper, we introduce an uncertainty decomposition framework for LLMs, called input clarifications ensemble, which bypasses the need to train new models. Rather than ensembling models with different parameters, our approach generates a set of clarifications for the input, feeds them into the fixed LLMs, and ensembles the corresponding predictions. We show that our framework shares a symmetric decomposition structure with BNN. Empirical evaluations demonstrate that the proposed framework provides accurate and reliable uncertainty quantification on various tasks. Code will be made publicly available at https://github.com/UCSB-NLP-Chang/llm_uncertainty .

翻译：不确定性分解是指将模型总不确定性分解为数据（随机）不确定性和模型（认知）不确定性的任务，前者源于数据固有的复杂性或模糊性，后者源于模型知识的不足。对大型语言模型（LLMs）进行不确定性分解是提升其可靠性、可信度和可解释性的重要步骤，但这一研究任务极具挑战性且尚未得到解决。现有经典方法——贝叶斯神经网络（BNN）无法应用于LLMs，因为BNN需要训练并集成多个模型变体，这对LLMs而言不可行或成本过高。本文提出一种适用于LLMs的不确定性分解框架，称为输入澄清集成，该方法无需训练新模型。与集成不同参数模型不同，我们的方法为输入生成一组澄清信息，将其输入固定的LLMs，并集成相应的预测结果。我们证明该框架与BNN具有对称的分解结构。实验评估表明，该框架在各种任务上能提供准确且可靠的不确定性量化。代码将于https://github.com/UCSB-NLP-Chang/llm_uncertainty 公开。