Hong Kong University of Science and Technology

Against Political Polarization: A Unified Framework for Tracing Evolving Political Ideologies on Social Media

The rapid growth of social media has greatly influenced political discourse, highlighting the need to understand individual political ideologies and their temporal dynamics. This task faces challenges such as data scarcity, abundant non-political content, costly and bias-prone ma...

The institutional politics of development finance: ODA’s role in shaping subnational DFI investments

Abstract Development finance institutions (DFIs), operated by donor governments to support private sector development in the Global South, often face informational barriers when deciding where to invest abroad. We argue that traditional Official Development Assistance (ODA), deli...

Multi-objective and cross-scale inverse design of temperature-control materials via physics-constrained machine learning

Abstract Temperature-control materials, notably composite phase-change materials (CPCMs), show great potential for the thermal management of next-generation power electronics. However, the synergistic optimization of multiple performance metrics still relies heavily on Edisonian...

Surface-Charge-Regulated MoS2 Membrane for Induced Agglomeration and Efficient Removal of Nanopollutants

Abstract Conventional membranes for nanopollutant removal are constrained by the permeability–selectivity trade-off inherent in size-exclusion mechanisms. Here, we report a surface-charge-regulated MoS2 membrane engineered by anchoring copper nanoparticles and quaternary ammonium...

Conformal network link prediction with false discovery rate control under unstructured missingness

We propose a new method for predicting multiple missing links in partially observed networks while controlling the false discovery rate (FDR), a largely unresolved challenge in network analysis. The main difficulty lies in handling complex dependencies and unknown missing pattern...

Research engagement profiles among Chinese undergraduate and postgraduate students

Research engagement is increasingly recognized as an important component of student development in higher education, particularly in medical and health sciences disciplines, but it is often measured through isolated activities or publication status alone. This study examined rese...

Collaborative Robots in Construction: A Proactivity Perspective

Abstract Achieving effective human–robot collaboration (HRC) in dynamic environments like construction requires robots to function as proactive peers, anticipating human intents and acting in advance. While current literature analyzes enabling technologies through a sense–plan–ac...

Peak Shaving and Valley Filling: Reconstructing Cognitive Load in Cultural Heritage VR Through Contextualized Scaffolding

Virtual reality supports immersive cultural heritage communication, yet engagement outpaces understanding. A structural cause is that information about an object is scattered across transient carriers (NPC dialogue, annotations, triggered videos), requiring working-memory integra...

An explainable biomedical foundation model via large-scale concept-enhanced vision–language pretraining

Artificial intelligence for medical imaging is required to be accurate and interpretable to clinicians. However, current multimodal biomedical foundation models often prioritize performance over explainability. Here we present ConceptCLIP, an explainable biomedical foundation mod...

Accuracy assessment of Chinese large language models in psoriasis management: a multicenter expert evaluation study

Psoriasis patients in China face severe barriers to disease knowledge acquisition and limited access to medical resources, creating an urgent demand for reliable, evidence-based patient education tools. This multicenter expert evaluation study aimed to systematically assess the q...

First-Class Constrained Types: Elaboration, Type Inference, Approximation, and a Characterization of Termination

We study a first-class treatment of constrained types, which were previously confined mostly to ML-style polymorphism. We define System FCCT, an extension of System F with polymorphic subtyping and constraint abstraction in types. A value of type c ⇒ τ can be used at type τ in an...

EngageCue: Enabling Balanced Engagement in Face-to-Face Group Collaboration via AI Cues Through Smart Glasses

In face-to-face group collaboration, uneven engagement often undermines efficiency and interaction quality. Although AI and wearable devices are increasingly applied to support collaboration, existing approaches largely remain confined to online or screen-based environments, leav...

Automated In-InjectorPort Conversion of AristolochicAcids to Aristolic Acids for Enhanced HPLC Fluorescence Detection

Abstract Aristolochic acids (AAs) and the herbs containing them are recognized human carcinogens. Emerging evidence indicates that AAs are also food and environmental contaminants, posing significant risks to human health. This study developed a highly sensitive AA detection meth...

Enhanced Antibiotics Sieving by Exfoliated TiS2 Membranes via Surface Functionalization and Passivation

Abstract Transition metal dichalcogenide (TMD) nanolaminate membranes hold great promise for molecular sieving due to their two-dimensional capillary structures, which facilitate size-restricted diffusion. However, many transition metal sulfides exhibit intrinsic instability and...

State entrepreneurialism under economic transition: Bounded rationality in China’s high-speed rail new town governance

Drawing on 16 high-speed rail (HSR) new towns along the Beijing–Shanghai corridor, this paper develops a middle-range explanation of how state entrepreneurialism is practiced and differentiated under conditions of economic transition in China. While the notion of state entreprene...

Empowering PolymericMaterials Discovery by ArtificialIntelligence

Abstract Polymeric materials underpin modern technologies spanning energy storage, microelectronics, healthcare and sustainable manufacturing. Yet their rational design remains exceptionally challenging because material performance emerges from complex interactions among molecula...

Cavity-Size-Dependent Recognition of Naphthalene Derivatives by 3,3′-Bipyridinium Tetracationic Cyclophanes

The recognition behavior of tetracationic cyclophanes is strongly influenced by the geometric match between the host cavity and aromatic guests. Herein, we compare two 3,3′-bipyridinium-based tetracationic cyclophanes, BPy-Box⁴⁺ and NBox⁴⁺, toward a series of naphthalene derivati...

Reinforcement Learning in the Era of Large Language Models: Challenges and Opportunities

Reinforcement learning (RL), is becoming essential in the post-training of large language models (LLMs), enhancing their capabilities and alignment with human preferences. However, adapting conventional RL to LLMs introduces challenges stemming from their massive parameter size a...

A multi-channel signal-enhanced fractional-order vibrational resonance coupled stochastic resonance array method for early fault detection of rolling bearings

In the field of early fault detection for rolling bearings, stochastic resonance (SR) and vibrational resonance (VR) have been widely adopted as effective nonlinear signal processing tools. Nonetheless, when it comes to enhancing weak fault characteristics, most existing methods...

Reliable Battery-free Backscatter Video Transmission for Ubiquitous Vision-based Sensing

Past decades have witnessed the rapid development of battery-free backscatter-based cameras, opening up promising long-term and sustainable surveillance applications by harvesting wireless energy. However, limited by the fluctuations of the ambient energy, existing backscatter sy...