Large Language Models (LLMs) have achieved significant breakthroughs across various domains, but they can still produce unreliable or misleading outputs. For responsible LLM applications, uncertainty quantification techniques are used to estimate a model's uncertainty about its outputs, indicating the likelihood that those outputs may be problematic. For LLM reasoning tasks, it is essential to estimate uncertainty not only in the final answer but also in the intermediate reasoning process, particularly to identify where uncertainty arises. Such information may enable more fine-grained and targeted interventions during inference. In this study, we investigate which metrics can effectively localize uncertain places within an LLM reasoning trajectory. Our study reveals that uncertain intermediate continuations are more likely to occur at tokens that are highly sensitive to perturbations in the embeddings of preceding tokens. In our experiments, we show that such perturbation-based metrics achieve stronger performance in localizing uncertain intermediate steps than baseline methods, including probability-based, sampling-based, and Bayesian-based approaches. Meanwhile, our proposed metrics also enjoy good simplicity and efficiency.
Adaptive Uncertainty-Aware Tree Search for Robust Reasoning
Institute of Software Chinese Academy of Sciences, Beijing, China（中国科学院软件研究所）；University of the Chinese Academy of Sciences（中国科学院大学）；Multimodal Experiences Lab, Dolby Laboratories Inc（多模态体验实验室，Dolby Laboratories Inc）；Institute of Artificial Intelligence and Robotics, Xi’an Jiaotong University（人工智能与机器人研究所，西安交通大学）；Electronic Information Science and Technology, Central South University（电子信息科学与技术，中南大学）
Weakly-supervised temporal action localization aims to learn detecting temporal intervals of action classes with only video-level labels. To this end, it is crucial to separate frames of action classes from the background frames (i.e., frames not belonging to any action classes). In this paper, we present a new perspective on background frames where they are modeled as out-of-distribution samples regarding their inconsistency. Then, background frames can be detected by estimating the probability of each frame being out-of-distribution, known as uncertainty, but it is infeasible to directly lea…
Data Science Section, IT University of Copenhagen（丹麦哥本哈根技术大学数据科学部门）；MaiNLP, Center for Information and Language Processing, LMU Munich（慕尼黑大学信息与语言处理中心MaiNLP）；Department of Applied Mathematics and Computer Science, Technical University of Denmark（丹麦技术大学应用数学与计算机科学系）；Pioneer Centre for Artificial Intelligence（先锋人工智能中心）；Munich Center for Machine Learning (MCML)（慕尼黑机器学习中心（MCML））
Recent advances in domain adaptation establish that requiring a low risk on the source domain and equal feature marginals degrade the adaptation's performance. At the same time, empirical evidence shows that incorporating an unsupervised target domain term that pushes decision boundaries away from the high-density regions, along with relaxed alignment, improves adaptation. In this paper, we theoretically justify such observations via a new bound on the target risk, and we connect two notions of relaxation for divergence, namely $β-$relaxed divergences and localization. This connection allows u…
Graz University of Technology（格拉茨理工大学）；Medical University of Graz（格拉茨医科大学）；The University of Auckland（奥克兰大学）
In landmark localization, due to ambiguities in defining their exact position, landmark annotations may suffer from large observer variabilities, which result in uncertain annotations. To model the annotation ambiguities of the training dataset, we propose to learn anisotropic Gaussian parameters modeling the shape of the target heatmap during optimization. Furthermore, our method models the prediction uncertainty of individual samples by fitting anisotropic Gaussian functions to the predicted heatmaps during inference. Besides state-of-the-art results, our experiments on datasets of hand radi…
GeoReasoner: Geo-localization with Reasoning in Street Views using a Large Vision-Language Model
The Hong Kong University of Science and Technology (Guangzhou)（香港科技大学（广州））；Tongji University（同济大学）；Independent Researcher（独立研究者）；Information Engineering University（信息工程大学）
Robust localization in dense urban scenarios using a low-cost sensor setup and sparse HD maps is highly relevant for the current advances in autonomous driving, but remains a challenging topic in research. We present a novel monocular localization approach based on a sliding-window pose graph that leverages predicted uncertainties for increased precision and robustness against challenging scenarios and per frame failures. To this end, we propose an efficient multi-task uncertainty-aware perception module, which covers semantic segmentation, as well as bounding box detection, to enable the loca…
Automatic anatomical landmark localization has made great strides by leveraging deep learning methods in recent years. The ability to quantify the uncertainty of these predictions is a vital component needed for these methods to be adopted in clinical settings, where it is imperative that erroneous predictions are caught and corrected. We propose Quantile Binning, a data-driven method to categorize predictions by uncertainty with estimated error bounds. Our framework can be applied to any continuous uncertainty measure, allowing straightforward identification of the best subset of predictions …
FERA: Uncertainty-Aware Federated Reasoning for Large Language Models
Indiana University（印第安纳大学）；The University of New South Wales（新南威尔士大学）；The Chinese University of Hong Kong（香港中文大学）；University of California San Diego（加州大学圣地亚哥分校）；Adobe Research（Adobe研究院）
In this paper, we investigate visual-based camera re-localization with neural networks for robotics and autonomous vehicles applications. Our solution is a CNN-based algorithm which predicts camera pose (3D translation and 3D rotation) directly from a single image. It also provides an uncertainty estimate of the pose. Pose and uncertainty are learned together with a single loss function and are fused at test time with an EKF. Furthermore, we propose a new fully convolutional architecture, named CoordiNet, designed to embed some of the scene geometry. Our framework outperforms comparable method…
Ultrasonic guided waves are commonly used to localize structural damage in infrastructures such as buildings, airplanes, bridges. Damage localization can be viewed as an inverse problem. Physical model based techniques are popular for guided wave based damage localization. The performance of these techniques depend on the degree of faithfulness with which the physical model describes wave propagation. External factors such as environmental variations and random noise are a source of uncertainty in wave propagation. The physical modeling of uncertainty in an inverse problem is still a challengi…
King Abdulllah University of Science and Technology (KAUST)（阿卜杜拉国王科技大学）；King Fahd University of Petroleum and Minerals (KFUPM)（法赫德国王石油与矿产大学）
Several techniques have been proposed over the years for automatic hypocenter localization. While those techniques have pros and cons that trade-off computational efficiency and the susceptibility of getting trapped in local minima, an alternate approach is needed that allows robust localization performance and holds the potential to make the elusive goal of real-time microseismic monitoring possible. Physics-informed neural networks (PINNs) have appeared on the scene as a flexible and versatile framework for solving partial differential equations (PDEs) along with the associated initial or bo…
Active, non-parametric peak detection is considered. As a use case, active source localization is examined and an uncertainty-based sampling scheme algorithm to effectively localize the peak from a few energy measurements is designed. It is shown that under very mild conditions, the source localization error with $m$ actively chosen energy measurements scales as $O(\log^2 m/m)$. Numerically, it is shown that in low-sample regimes, the proposed method enjoys superior performance on several types of data and outperforms the state-of-the-art passive source localization approaches and in the low s…
Concept-Grounded Reasoning with Prompt-Driven Localization for Interpretable Structured Report Generation
The Hong Kong University of Science and Technology（香港科技大学）；The Chinese University of Hong Kong（香港中文大学）；The Hong Kong University of Science and Technology (Guangzhou)（香港科技大学（广州））；Saarland University（萨尔兰大学）；Mohamed bin Zayed University of Artificial Intelligence（穆罕默德·本·扎耶德人工智能大学）；University of Cambridge（剑桥大学）
University of Pennsylvania（宾夕法尼亚大学）；New Jersey Institute of Technology（新泽西理工学院）；Washington University in St. Louis（圣路易斯华盛顿大学）；Argonne National Laboratory（阿贡国家实验室）
