As AI systems increasingly process video, audio, and language data, reducing unnecessary computation has become a requirement. Researchers from the Japan Advanced Institute of Science and Technology (JAIST) developed an AI model that identifies important video segments and processes only relevant information. The model reduced processing time for 2-minute clips by 65%, from 52 to 18 seconds, while maintaining high accuracy. This approach could enable efficient, real-time AI coaching and communication tools on everyday devices.

Before their first semester at Rice University begins, 38 incoming students are already getting a taste of what it means to be a Rice student: tackling demanding STEM courses, learning how to study collaboratively, meeting faculty and building a community they can rely on when classes begin in the fall. They are part of the Rice Emerging Scholars Program, or RESP, an intensive residential program designed to prepare first-year students for the pace, depth and rigor of STEM education at Rice.

The explosion of artificial intelligence (AI) and the era of Big Data have exposed a critical flaw in traditional computer hardware: the slow and energy-consuming data transfer between memory and processors, commonly known as the von Neumann bottleneck. Every day, when our devices process AI queries, they must move massive amounts of data back and forth, causing severe delays and battery drain.

To address this challenge, researchers at The University of Hong Kong (HKU) have created a chip with a minimalist architecture that performs complex searches directly where the data is stored, fundamentally bypassing this bottleneck. The research team, led by Professor Can Li from the Department of Electrical and Computer Engineering of the Faculty of Engineering and the Centre for Advanced Semiconductors and Integrated Circuits (CASIC), has developed a novel analogue content-addressable memory (CAM) utilising two-dimensional (2D) MoS2 flash memories.

Professor Hengshuang Zhao of the School of Computing and Data Science at The University of Hong Kong (HKU) has been selected as one of the 2025 "35 Innovators Under 35 China" (TR35 China) by MIT Technology Review, a technology publication founded by the Massachusetts Institute of Technology (MIT) in the United States.

The Smart QS Hackathon 2026, co-organised for the first time by the Housing Bureau and the Hong Kong Housing Authority (HA), Cyberport, and the University of Hong Kong (HKU), was officially launched today (July 29) with a welcome ceremony and briefing session held at Cyberport. The Hackathon aims to explore innovative applications of Artificial Intelligence (AI) in Quantity Surveying (QS) through cross-sector collaboration, facilitating the intelligent transformation of the construction industry and enhancing the management and service effectiveness of public works.

A research team led by Professor Ruibang LUO, Assistant Director (Learning Experience & Student Enrichment) and Associate Head of the Department of AI & Data Science, School of Computing and Data Science (CDS), The University of Hong Kong (HKU), has developed ClairS—a novel deep-learning algorithm that significantly improves the detection of cancer mutations using long-read sequencing. Tested on breast cancer, lung cancer and melanoma cell-line datasets, ClairS has demonstrated exceptional accuracy across various cancer types and sequencing conditions.

Using simple artificial cells, a research group led have uncovered fundamental physical principles that govern how living cells change their shape. By combining experiments with computer simulations and theoretical analysis, the researchers showed how a cell’s actin cytoskeleton—the network of protein fibers that gives it is shape—can generate cell-scale shape changes and front-rear polarity without relying on complex biochemical signaling.