可信概念瓶颈模型用于知识-随机不确定性分解

HybridFlow：使用单一混合模型量化偶然不确定性和认知不确定性

HybridFlow: Quantification of Aleatoric and Epistemic Uncertainty with a Single Hybrid Model

HybridFlow提出一种模块化混合架构，结合条件掩码自回归流与概率预测器，统一量化偶然和认知不确定性，在回归任务中提升校准性并更好对齐模型误差。

可信概念瓶颈模型：epistemic和aleatoric不确定性的结构分离

Structurally Separated Uncertainty in Supervised Latent Variable Models

本文提出了一种可信概念瓶颈模型，通过结构分离epistemic和aleatoric不确定性，显著降低两者相关性并提升预测误差与真实模糊度的一致性。

Learning to Forecast Aleatoric and Epistemic Uncertainties over Long Horizon Trajectories

Giving autonomous agents the ability to forecast their own outcomes and uncertainty will allow them to communicate their competencies and be used more safely. We accomplish this by using a learned world model of the agent system to forecast full agent trajectories over long time horizons. Real world systems involve significant sources of both aleatoric and epistemic uncertainty that compound and interact over time in the trajectory forecasts. We develop a deep generative world model that quantifies aleatoric uncertainty while incorporating the effects of epistemic uncertainty during the learni…

用于不确定性量化的Credal集成蒸馏

Credal Ensemble Distillation for Uncertainty Quantification

针对深度集成推理成本高的问题，提出Credal集成蒸馏（CED）框架，将深度集成压缩为单模型CREDIT，通过预测类概率区间定义Credal集实现不确定性量化，在分布外检测基准上表现优异且推理开销更低。

对任意性的认知与偶然分解以约束优良模型集合

An Epistemic and Aleatoric Decomposition of Arbitrariness to Constrain the Set of Good Models

本研究将机器学习模型的不稳定性分解为认知和偶然两部分，揭示了偶然不稳定性在不同群体间的显著差异，并提出一种结合准确性与公平性的模型选择方法，有效筛选出稳定、公平且准确的模型。

Evaluating aleatoric and epistemic uncertainties of time series deep learning models for soil moisture predictions

Soil moisture is an important variable that determines floods, vegetation health, agriculture productivity, and land surface feedbacks to the atmosphere, etc. Accurately modeling soil moisture has important implications in both weather and climate models. The recently available satellite-based observations give us a unique opportunity to build data-driven models to predict soil moisture instead of using land surface models, but previously there was no uncertainty estimate. We tested Monte Carlo dropout (MCD) with an aleatoric term for our long short-term memory models for this problem, and ask…

物理信息神经网络中偶然不确定性与认知不确定性的解耦：绝缘材料退化预测应用

Disentangling Aleatoric and Epistemic Uncertainty in Physics-Informed Neural Networks. Application to Insulation Material Degradation Prognostics

提出异方差贝叶斯物理信息神经网络（B-PINN），联合建模认知与偶然不确定性，用于时空绝缘材料老化预测，在变压器绝缘老化应用中比确定性PINN和dropout-PINN具有更高精度和校准的不确定性估计。

CUPID：一种联合估计自然不确定性和本质不确定性的插件框架

CUPID: A Plug-in Framework for Joint Aleatoric and Epistemic Uncertainty Estimation with a Single Model

CUPID是一种无需修改基础模型即可联合估计自然和本质不确定性的插件框架，适用于多种任务，提供透明且可解释的不确定性分析。

从答案到解释：重新思考大语言模型（LLM）中歧义诱导的偶然不确定性估计

From Answers to Interpretations: Rethinking Ambiguity-Induced Aleatoric Uncertainty Estimation in LLMs

该研究针对LLM部署中歧义诱导的偶然不确定性估计问题，提出仅基于解释空间的直接方法，提升了AUROC并降低了计算成本，其效果优于现有依赖答案的方法。

UbiQTree: 在XAI中利用树集成进行不确定性量化

UbiQTree: Uncertainty Quantification in XAI with Tree Ensembles

UbiQTree通过整合Dempster-Shafer理论和Dirichlet过程，分解SHAP值中的不确定性，提升XAI的可靠性与解释性。

当然不确定：面向多模态认知不确定性与偶然不确定性感知的基准与指标

Certainly Uncertain: A Benchmark and Metric for Multimodal Epistemic and Aleatoric Awareness

针对视觉语言 AI 的不确定性感知，本文区分认知与偶然不确定性，构建含 17.8 万 VQA 对比样本的 CertainlyUncertain 基准，并提出置信度加权准确率指标。

在认知不确定性下推广统计学习理论中的可实现性

Generalising realisability in statistical learning theory under epistemic uncertainty

本文研究当训练与测试分布来自同一credal集（概率分布凸集）时，统计学习理论中可实现性等核心概念的推广问题，为认知不确定性下的统计学习奠定理论基础。

从偶然不确定性到认知不确定性：探索人工智能中的不确定性量化技术

From Aleatoric to Epistemic: Exploring Uncertainty Quantification Techniques in Artificial Intelligence

本文综述AI中偶然与认知不确定性的量化方法，梳理概率、集成、采样、生成及混合技术，并讨论应用、挑战与未来方向。

Probabilities Are Not Enough: Formal Controller Synthesis for Stochastic Dynamical Models with Epistemic Uncertainty

Capturing uncertainty in models of complex dynamical systems is crucial to designing safe controllers. Stochastic noise causes aleatoric uncertainty, whereas imprecise knowledge of model parameters leads to epistemic uncertainty. Several approaches use formal abstractions to synthesize policies that satisfy temporal specifications related to safety and reachability. However, the underlying models exclusively capture aleatoric but not epistemic uncertainty, and thus require that model parameters are known precisely. Our contribution to overcoming this restriction is a novel abstraction-based co…

模型置信度对视觉-语言模型测量不确定性中偏差效应的影响

The Role of Model Confidence on Bias Effects in Measured Uncertainties for Vision-Language Models

本研究通过VQA实验发现，缓解提示偏差可改善GPT-4o的不确定性量化，且无偏差置信度越低，偏差对认知和偶然不确定性的影响越大，并导致认知不确定性被低估和过度自信。

重新思考主动开集标注中的认知不确定性与任意不确定性：一种基于能量的方法

Rethinking Epistemic and Aleatoric Uncertainty for Active Open-Set Annotation: An Energy-Based Approach

针对开集主动学习中认知与任意不确定性利用不当的问题，提出 EAOA 框架，通过能量度量、检测器损失和目标驱动自适应采样实现最先进标注性能。

XU-RS：解释随机集语言模型中的信度宽度

XU-RS: Explaining Credal Width in Random-Set Language Models

本文提出XU-RS框架，利用期望梯度归因方法解释随机集语言模型中答案信度宽度的来源，并在MedQA数据集上验证了其有效性，同时探讨了归一化影响与诊断检查。

用于分类中不确定性估计的模型平均的信度包装方法

Credal Wrapper of Model Averaging for Uncertainty Estimation in Classification

本文提出信度包装方法，将BNN和深度集成的模型平均表示为信度集，通过类别概率上下界和交集概率变换改进分类不确定性估计，并在多个OOD检测基准上验证其优于基线且校准误差更低。

证据深度学习方法能否忠实地表示认知不确定性？

Is Epistemic Uncertainty Faithfully Represented by Evidential Deep Learning Methods?

本文从理论上分析证据深度学习方法，揭示其二阶损失优化、可辨识性、收敛性及认知不确定性度量仅具相对意义等根本问题。

不变变换与重采样基于的信念不确定性减少

Invariant Transformation and Resampling based Epistemic-Uncertainty Reduction

本文提出基于不变变换和重采样的方法，用于减少信念不确定性，提高AI模型的推理准确性。

一种用于评估基于集合的认知不确定性表示的校准测试

A calibration test for evaluating set-based epistemic uncertainty representations

该研究针对基于集合的认知不确定性表示的有效性验证问题，提出一种依赖实例的凸组合校准统计检验方法，通过恰当评分规则与核估计器实现非参数检验，在合成和真实实验中验证了捕捉实例可变性的优势。

可解释不确定性路由：分离面部表情识别中的情感歧义与分布偏移

Interpretable Uncertainty Routing Separating Emotion Ambiguity from Distribution Shift in Facial Expression Recognition

提出不确定性感知路由（UAR），通过将不确定性分解为偶然不确定性和认知不确定性，分别对应标注者分歧和分布偏移，实现歧义样本报告与异常样本拒绝的差异化处理。

认知性拒绝选项预测

Epistemic Reject Option Prediction

本文提出认知性拒绝选项预测器，基于贝叶斯学习以最小化期望遗憾为目标，在数据不足导致的高认知不确定性区域弃权，首次为识别训练数据不足以可靠决策的输入提供了原则性框架。

ReflectiChain: 面向供应链韧性的LLM驱动世界模型中的认知基础

ReflectiChain: Epistemic Grounding in LLM-Driven World Models for Supply Chain Resilience

提出ReflectiChain框架，通过生成式供应链世界模型和双环学习分离认知不确定性与偶然不确定性，在半导体基准上提升推理一致性33.0%，并在对抗冲击下保持82.3%可操作性。

Enabling risk-aware Reinforcement Learning for medical interventions through uncertainty decomposition

Reinforcement Learning (RL) is emerging as tool for tackling complex control and decision-making problems. However, in high-risk environments such as healthcare, manufacturing, automotive or aerospace, it is often challenging to bridge the gap between an apparently optimal policy learnt by an agent and its real-world deployment, due to the uncertainties and risk associated with it. Broadly speaking RL agents face two kinds of uncertainty, 1. aleatoric uncertainty, which reflects randomness or noise in the dynamics of the world, and 2. epistemic uncertainty, which reflects the bounded knowledge…

用于不确定性下语义承诺的Credal大语言模型

Credal Large Language Models for Semantic Commitment under Uncertainty

本研究提出Credal大语言模型（CLLMs），通过LoRA适配器集成体构建Credal集合，推导CTC和SCC两类承诺分数，在多模型多数据集上验证其在问答、幻觉检测等任务中表现优异。

何处测量：基于epistemic不确定性的传感器布置与ConvCNPs

Where to Measure: Epistemic Uncertainty-Based Sensor Placement with ConvCNPs

本文提出基于epistemic不确定性的传感器布置方法，通过扩展ConvCNPs并引入MDNs输出头，更有效地减少模型误差。

随机策略-批评法：通过时间性随机不确定性缓解过估计

Stochastic Actor-Critic: Mitigating Overestimation via Temporal Aleatoric Uncertainty

STAC通过引入时间性随机不确定性来缓解强化学习中批评器的过估计问题，采用分布性批评器和dropout正则化提升稳定性和效率。

集成分位网络：面向自动驾驶应用的不确定性感知强化学习

Ensemble Quantile Networks: Uncertainty-Aware Reinforcement Learning with Applications in Autonomous Driving

本文提出集成分位网络（EQN），结合分布强化学习与集成方法同时估计偶然和认知不确定性，使自动驾驶智能体能在遮挡交叉口平衡风险与效率，并识别分布外情境以避免危险决策。

评估认知不确定性：超越分布外检测和主动学习

Evaluating Epistemic Uncertainty: Beyond OOD Detection and Active Learning

研究认知不确定性评估，基于认知拒绝选项框架，通过将选择性预测表述为约束优化证明最优选择器，揭示现有文献弱点，提议评估分解的可实现风险等作为诊断，实验表明决策理论排名与代理任务排名有差异。

认知深度学习：使机器学习模型能够知道自己不知道什么

Epistemic Deep Learning: Enabling Machine Learning Models to Know When They Do Not Know

本论文提出认知深度学习范式，通过随机集神经网络量化认知不确定性，使模型在不确定时弃权（不执行），并适配大语言模型及自动驾驶天气分类，经统一评估框架验证，提升可靠性。

用专门的认知不确定性模型重新设计集合卡尔曼滤波器

Redesigning the ensemble Kalman filter with a dedicated model of epistemic uncertainty

针对实际问题中认知不确定性难以概率建模的问题，本文提出可能性集合卡尔曼滤波器，基于可能性理论刻画认知不确定性，在小样本量下表现良好且在给定样本量时优于标准方法。

深度回归中变量误差模型的偶然不确定性

Aleatoric uncertainty for Errors-in-Variables models in deep regression

本文提出将变量误差概念引入贝叶斯深度回归，以量化输入不确定性，实现更完整的偶然与认知不确定性分解，在不损失预测性能的同时提升不确定性估计的可靠性。

量化扩散模型的先验不确定性

Quantifying Epistemic Uncertainty in Diffusion Models

本文提出FLARE方法，通过基于信息量的随机近似技术，有效分离扩散模型的先验不确定性，提升生成数据的可信度评估。

ConformaDecompose：通过校准局部化解释不确定性

ConformaDecompose: Explaining Uncertainty via Calibration Localization

本文提出一种不确定性意识的可解释框架，通过逐步校准局部化分析回归任务中校准诱导的epistemic conformal不确定性可减少性，揭示不确定性来源，增强可解释性。

Decision-Making Under Uncertainty: Beyond Probabilities

This position paper reflects on the state-of-the-art in decision-making under uncertainty. A classical assumption is that probabilities can sufficiently capture all uncertainty in a system. In this paper, the focus is on the uncertainty that goes beyond this classical interpretation, particularly by employing a clear distinction between aleatoric and epistemic uncertainty. The paper features an overview of Markov decision processes (MDPs) and extensions to account for partial observability and adversarial behavior. These models sufficiently capture aleatoric uncertainty but fail to account for…

基于CBAM增强的EfficientNet和证据深度学习的不确定性意识卫星图像野火烟密度分类

Uncertainty-Aware Wildfire Smoke Density Classification from Satellite Imagery via CBAM-Augmented EfficientNet with Evidential Deep Learning

本文提出一种概率框架，通过CBAM增强的EfficientNet和证据深度学习，对卫星图像中的烟雾密度进行分类，并提供分解的epistemic和aleatoric不确定性。模型在16298个真实卫星图像块上达到93.8%的加权测试准确率。

深度集成用于知识不确定性：一个频数视角

Deep Ensembles for Epistemic Uncertainty: A Frequentist Perspective

本文从频数视角探讨深度集成在量化epistemic不确定性中的有效性，通过bootstrap估计器和数据变异性的分解，揭示了深度集成捕捉训练随机性成分的机制。

SPUQ：面向大语言模型的基于扰动的不确定性量化

SPUQ: Perturbation-Based Uncertainty Quantification for Large Language Models

针对LLM高置信错误预测及认知不确定性研究不足的问题，提出SPUQ扰动采样与聚合方法，显著改善不确定性校准，ECE平均降低50%。

主观风险分解：不确定性量化的新视角

Subjective Risk Decomposition: A New View for Uncertainty Quantification

该研究提出不确定性量化新观点，基于严格恰当损失通过主观风险分解推导认知和偶然不确定性度量，恢复众多UQ度量并提供理论基础，还扩展到学习理论，为不确定性量化完整学习理论框架迈出第一步。

Bayesian Deep Learning for Affordance Segmentation in images

Affordances are a fundamental concept in robotics since they relate available actions for an agent depending on its sensory-motor capabilities and the environment. We present a novel Bayesian deep network to detect affordances in images, at the same time that we quantify the distribution of the aleatoric and epistemic variance at the spatial level. We adapt the Mask-RCNN architecture to learn a probabilistic representation using Monte Carlo dropout. Our results outperform the state-of-the-art of deterministic networks. We attribute this improvement to a better probabilistic feature space repre…

定量超声中Homodyned-K分布的不确定性分解与误差裕度检测

Uncertainty Decomposition and Error Margin Detection of Homodyned-K Distribution in Quantitative Ultrasound

本研究提出一种方法，对BNN估计的定量超声HK分布参数在仿真与实验数据中的认知和偶然不确定性进行分解，并探究预测误差与两类不确定性的关联，为临床判断QUS特征值可信度提供依据。