This study delves into the burgeoning domain of Generative Artificial Intelligence (GAI)
within the context of Industry 5.0 (I-5.0), highlighting the pivotal role of advanced GAI models …

Recent Large Language Models (LLMs) have shown the ability to generate content that is
difficult or impossible to distinguish from human writing. We investigate the ability of …

Constructing fast samplers for unconditional diffusion and flow-matching models has
received much attention recently; however, existing methods for solving inverse problems …

In climate science and meteorology, high-resolution local precipitation (rain and snowfall)
predictions are limited by the computational costs of simulation-based methods. Statistical …

Visual diffusion models have revolutionized the field of creative AI, producing high-quality
and diverse content. However, they inevitably memorize training images or videos …

Abstract Large Language Models are known to capture real-world knowledge, allowing
them to excel in many downstream tasks. Despite recent advances, these models are still …

To explore threats and opportunities and to chart a path for safely navigating the rapid
changes that generative artificial intelligence (AI) will bring to clinical research, the Duke …

Machine learning (ML) has the potential to improve various supply chain management
(SCM) tasks, namely demand forecasting, risk management, inventory management …

There is a lack of quantitative measures to evaluate the progression of topics through time in
dynamic topic models (DTMs). Filling this gap, we propose a novel evaluation measure for …

Deep learning models need a sufficient amount of data in order to be able to find the hidden
patterns in it. It is the purpose of generative modeling to learn the data distribution, thus …