News & Stories

人工智能（AI）圖像編輯及生成模型獲廣泛應用於圖像創作，然而其對抽象概念如感覺和氛圍等理解精準度一直存在局限，且多依賴純文字指令，較難準確表達複雜圖像意思，亦無法捕捉風格、材質或光影等效果。由香港科技大學（科大）馮諾依曼研究院院長兼計算機科學及工程學系講座教授賈佳亞教授帶領的團隊成功開發名為「DreamOmni2」的AI圖像生成和編輯器，不僅擁有卓越的多模態指令編輯和實體物件生成能力，更在抽象概念的理解和生成方面有重大突破，讓AI不僅能「看圖」，更能「理解圖意」，多方面表現優於同類型開源和閉源模型，為AI創作開啟無限可能。直擊缺陷：解鎖抽象概念的創作潛能近年，圖像編輯及生成模型的發展進入爆發階段，新品頻出，但暫時仍未有任何模型能徹底克服實際操作上遇到的兩大缺陷。其一是文本指令的局限性，純文字指令難以準確描述人物特徵、抽象紋理等細節。其二是抽象概念的缺失，現有模型僅能處理有形實體，如人物、物件，無法有效應對髮型、妝容、紋理、光影效果或風格等抽象概念，極大程度上限制了創作空間。DreamOmni2則可解決有關難題，成功執行兩大全新任務，包括根據用家輸入的抽象或實體概念，執行多模態指令編輯和生成，真正成為「全能創作工具」。全面性能測試：超越現有開源與閉源模型在多模態指令生成任務中，DreamOmni2能基於圖片中的實體進行圖像生成，例如提取圖一的畫作掛在臥室牆上，將圖二盤子的材質套用在圖三的水杯，並將水杯放置在桌子上，以生成符合用家要求的新圖像（圖示一）。在多模態指令編輯任務中，DreamOmni2的表現亦非常優秀，例如將圖中帽子的顏色變成與另一張圖毛衣相同的配色（圖示二）。

2025年10月16日至19日，香港科大「創科行」（第4站）＠HICOOL 2025全球創業家高峰會系列活動成功舉辦！HICOOL全球創業家高峰會被譽為「全球創新創業風向標」，本屆高峰會吸引100餘個國家和地區的頂尖創新力量齊聚。作為「創科行」活動的核心連結場景，本屆高峰會活動為香港科大搭建了「成果展示+思想碰撞」的雙重平台，成為本次創科行的重頭戲。

A landmark study conducted by The Hong Kong University of Science and Technology (HKUST) has demonstrated that a genetic variant, TREM2 H157Y, significantly increases the risk of Alzheimer’s disease (AD) in individuals of ethnic Chinese descent. The research reveals that this variant confers a risk level comparable to that of the strongest known genetic risk factor for AD, APOE-ε4, and is associated with more rapid clinical progression and more severe neurodegeneration.This is the first in the field to conduct an in-depth family-based clinical case study on a Chinese-enriched genetic risk factor for AD, the TREM2 H157Y variant. The findings, published in Alzheimer’s & Dementia: The Journal of the Alzheimer’s Association, have profound implications for disease monitoring and patient management.

The Hong Kong University of Science and Technology (HKUST), the Education Bureau (EDB), and the Chinese Academy of Sciences' Technology and Engineering Center for Space Utilization (CAS-TECSU) jointly organized the Hong Kong Secondary Student Space Station Popular Science Payload and Science Experiment Design Challenge. An award ceremony was held at HKUST on September 21 to commend the outstanding student teams for their excellent designs and recognize their achievements in space science.

The Hong Kong University of Science and Technology (HKUST) selected eight cutting-edge projects for the latest round of Sustainable Smart Campus as a Living Lab program for testing on campus. Covering a wide range of sustainability challenges on campus, the projects address energy efficiency, renewable energy optimization, water management, wellness, ecosystem enhancement, and predictive maintenance strategies, providing students and faculty with hands-on experience with prototypes before real-world implementation.

The Hong Kong University of Science and Technology (HKUST) today announced the launch of “Aivilization”, the largest-scale AI multi-agent social simulation sandbox to date. This innovative platform studies how human interaction with AI can shape a virtual society, where ‘AI inhabitants’ can organically develop their own governance structures, economies and cultural norms.   As AI becomes increasingly integrated into our daily lives, understanding the dynamics of communication between humans and artificial intelligence has become a crucial area of study. Previous simulation models like Project Sid1 and Stanford Smallville2 have explored similar themes, with community sizes of around 1,000 and dozens of AI agents respectively, to observe the evolution of AI civilizations.

The Hong Kong University of Science and Technology (HKUST) Institute for Emerging Market Studies, EY1 and Golden Age Foundation have jointly announced the signing of a Memorandum of Understanding (MoU) to officially activate the development of the “Smart Ecosystem for the Silver Population” in the Guangdong-Hong Kong-Macao Greater Bay Area (GBA). This initiative aims to promote the development of the region’s elderly care industry. The signing ceremony, held during the “BOC Life 10th Golden Age Expo and Summit”, was witnessed by Mr. Ho Kai Ming, JP, Under Secretary for Labour and Welfare of the Hong Kong Special Administrative Region. 

The buzz surrounding drones and eVTOL vehicles signals an exciting future—faster deliveries, rapid medical aid, and efficient public services just over our heads. But as these technologies take flight, they face two critical challenges: noise pollution and public safety. Drone and eVTOL operation in low altitude airspace produce noise and face flight safety challenges in micro-weather and building wind environment. Professor ZHANG Xin, Swire Professor of Aerospace Engineering, and Professor ZHOU Peng at The Hong Kong University of Science and Technology (HKUST) are leading efforts to overcome these issues. Combining cutting-edge research in aerospace engineering with practical solutions, their team is striving to ensure drones and eVOTL  vehicles can integrate harmoniously into urban life.