Sharing cutting-edge global technological trends, presenting research achievements, and fostering industry-academia collaboration, UNIST Graduate School of Artificial Intelligence hosted the ‘2025 UNIST AI Technology Open Workshop... Read More
Abstract Flat minima, known to enhance generalization and robustness in supervised learning, remain largely unexplored in generative models. In this work, we systematically investigate the role of loss surface flatness in generative models,... Read More
The School of New UNISTars hosted its annual major exploration festival, ‘2025 UNIVISION‘ in the lobby of the Engineering Building II on October 27–28, 2025. The two-day event brought together more than 200 freshmen and 6... Read More
Abstract Nitrous oxide (N2O) is one of the top three greenhouse gases, together with carbon dioxide and methane, but it is stable enough not to be easily decomposed under thermocatalytic conditions, even at high temperature (445 °C). Herein... Read More
Vacant retail stores in the heart of Ulsan are transforming into vibrant laboratories for imagining the city’s future. From September 25 to November 1, UNIST will host the ‘Public Design Laboratory 2025: Envisioning Policies for... Read More
Abstract Using the Surface Particulate Matter Network (SPARTAN) and Aerosol Robotic Network (AERONET), we investigated the relationship between aerosol chemical composition and optical properties across 14 global sites. The mass concentrati... Read More
UNIST has officially initiated a comprehensive fundraising campaign, “Vision 2050,” aimed at realizing its strategic growth blueprint for the mid- to long-term future. To mark the campaign’s launch, UNIST President Chong R... Read More
UNIST Novatus Graduate School and the Ulsan Regional SMEs and Startups Office have entered into a strategic partnership aimed at enhancing AI capabilities and cultivating talent among local small and medium-sized enterprises (SMEs). The sig... Read More
Abstract Achieving a comprehensive understanding of battery systems necessitates multi-length scale analysis, from the atomic- to macro-scale, to grasp the complex interplay of phenomena influencing performance. However, studies to understa... Read More
Abstract Understanding 3D scenes semantically and spatially is crucial for the safe navigation of robots and autonomous vehicles, aiding obstacle avoidance and accurate trajectory planning. Camera-based 3D semantic occupancy prediction, whi... Read More
![[UNIST Webinar] AI enabled Computational Imaging for Biomedical Applications](https://news.unist.ac.kr/wp-content/uploads/2022/10/Session-13.18-296x166.jpg)