English
Related papers

Related papers: Entropy Heat-Mapping: Localizing GPT-Based OCR Err…

200 papers

Over the last few years, machine learning unlocked previously infeasible features for compression, such as providing guarantees for users' privacy or tailoring compression to specific data statistics (e.g., satellite images or audio…

Information Theory · Computer Science 2026-03-25 Gergely Flamich

Vision-language models (VLMs) achieve remarkable performance but remain vulnerable to adversarial attacks. Entropy, as a measure of model uncertainty, is highly correlated with VLM reliability. While prior entropy-based attacks maximize…

Computer Vision and Pattern Recognition · Computer Science 2026-05-26 Mengqi He , Xinyu Tian , Xin Shen , Jinhong Ni , Shu Zou , Zhaoyuan Yang , Jing Zhang

Balanced and efficient information flow is essential for optimizing language generation models. In this work, we propose Entropy-UID, a new token selection method that balances entropy and Uniform Information Density (UID) principles for…

Computation and Language · Computer Science 2025-02-21 Xinpeng Shou

The proliferation of misleading visualizations online, particularly during critical events like public health crises and elections, poses a significant risk. This study investigates the capability of GPT-4 models (4V, 4o, and 4o mini) to…

Computer Vision and Pattern Recognition · Computer Science 2025-05-23 Jason Alexander , Priyal Nanda , Kai-Cheng Yang , Ali Sarvghad

Entropy minimization (EM) is frequently used to increase the accuracy of classification models when they're faced with new data at test time. EM is a self-supervised learning method that optimizes classifiers to assign even higher…

Computer Vision and Pattern Recognition · Computer Science 2024-05-14 Ori Press , Ravid Shwartz-Ziv , Yann LeCun , Matthias Bethge

Cold-start problem is a fundamental challenge for recommendation tasks. The recent self-supervised learning (SSL) on Graph Neural Networks (GNNs) model, PT-GNN, pre-trains the GNN model to reconstruct the cold-start embeddings and has shown…

Information Retrieval · Computer Science 2022-05-24 Bowen Hao , Hongzhi Yin , Jing Zhang , Cuiping Li , Hong Chen

In computer vision, we have the problem of creating graphs out of unstructured point-sets, i.e. the data graph. A common approach for this problem consists of building a triangulation which might not always lead to the best solution. Small…

Computer Vision and Pattern Recognition · Computer Science 2015-05-26 Samuel de Sousa , Walter G. Kropatsch

Entropy is a measure of self-information which is used to quantify losses. Entropy was developed in thermodynamics, but is also used to compare probabilities based on their deviating information content. Corresponding model uncertainty is…

Probability · Mathematics 2018-01-23 Alois Pichler , Ruben Schlotter

We prove an entropy versus degree dichotomy for low-degree tests and the Sum-of-Squares (SoS) hierarchy on a calibrated window after a gadget layer. For a target distribution \(\mu\) and a product-like proxy \(u\), we study the low-degree…

Computational Complexity · Computer Science 2025-09-30 Marko Lela

Imaging Inverse problems aim to reconstruct an underlying image from undersampled, coded, and noisy observations. Within the wide range of reconstruction frameworks, the unrolling algorithm is one of the most popular due to the synergistic…

Image and Video Processing · Electrical Eng. & Systems 2026-04-16 Roman Jacome , Romario Gualdrón-Hurtado , Leon Suarez-Rodriguez , Henry Arguello

Keypoint detection underpins many vision tasks, including pose estimation, viewpoint recovery, and 3D reconstruction, yet modern neural models remain vulnerable to small input perturbations. Despite its importance, formal robustness…

Computer Vision and Pattern Recognition · Computer Science 2026-03-09 Xusheng Luo , Changliu Liu

Recent reasoning Large Language Models (LLMs) demonstrate remarkable problem-solving abilities but often generate long thinking traces whose utility is unclear. Our work aims to improve their efficiency, enabling them to reach high…

Computation and Language · Computer Science 2026-05-11 Xiang Liu , Xuming Hu , Xiaowen Chu , Eunsol Choi

Vision-Language Models for remote sensing have shown promising uses thanks to their extensive pretraining. However, their conventional usage in zero-shot scene classification methods still involves dividing large images into patches and…

Computer Vision and Pattern Recognition · Computer Science 2025-01-08 Karim El Khoury , Maxime Zanella , Benoît Gérin , Tiffanie Godelaine , Benoît Macq , Saïd Mahmoudi , Christophe De Vleeschouwer , Ismail Ben Ayed

Mixture distributions are a workhorse model for multimodal data in information theory, signal processing, and machine learning. Yet even when each component density is simple, the differential entropy of the mixture is notoriously hard to…

Information Theory · Computer Science 2026-02-18 Namyoon Lee

Approximation of entropies of various types using machine learning (ML) regression methods are shown for the first time. The ML models presented in this study define the complexity of the short time series by approximating dissimilar…

Machine Learning · Computer Science 2022-11-30 Andrei Velichko , Maksim Belyaev , Matthias P. Wagner , Alireza Taravat

We present a validation of the asdf method, an information-theoretic framework for computing thermodynamic entropy from molecular configurations. The method reformulates entropy estimation as the Shannon entropy of a residual mapping…

Statistical Mechanics · Physics 2026-04-29 Dallin Fisher , Qi-Jun Hong

We propose the use of entropy, measured from the spatial and flux distribution of pixels in the residual image, as a potential diagnostic and stopping metric for the CLEAN algorithm. Despite its broad success as the standard deconvolution…

Instrumentation and Methods for Astrophysics · Physics 2023-11-10 D. C. Homan , J. S. Roth , A. B. Pushkarev

In this paper, we introduce a novel regularization scheme on contrastive language-image pre-trained (CLIP) medical vision models. Our approach is based on the observation that on many medical imaging tasks text tokens should only describe a…

Machine Learning · Computer Science 2023-03-01 Anil Palepu , Andrew L. Beam

Manuscript symbols can be stored, recognized and retrieved from an entropic digital memory that is associative and distributed but yet declarative; memory retrieval is a constructive operation, memory cues to objects not contained in the…

Machine Learning · Computer Science 2022-10-12 Rafael Morales , Noé Hernández , Ricardo Cruz , Victor D. Cruz , Luis A. Pineda

Global point cloud registration is essential in many robotics tasks like loop closing and relocalization. Unfortunately, the registration often suffers from the low overlap between point clouds, a frequent occurrence in practical…

Robotics · Computer Science 2023-07-25 Zhijian Qiao , Zehuan Yu , Huan Yin , Shaojie Shen
‹ Prev 1 8 9 10 Next ›