Junyu Lai

Toward Efficient Publish-Subscribe Data Dissemination in NDN-Based UAV Swarm Networks for Low-Altitude Internet

Haofang Dong, Junyu Lai, Heng Li, Yunfei Zhao et autres

cn, bg, be (code pays fourni par la source)

Optimizing Resource Utilization and Performance in LEO Satellite Edge Computing: A Joint Service Deployment and Task Offloading Approach

Junyu Lai, Di Wu, Yuhang Chen, Huashuo Liu et autres

While service-oriented low Earth orbit (LEO) satellite edge computing (LSEC) frameworks enable diverse edge services for user tasks, the heterogeneous distribution of terrestrial users causes substantial imbalance in computational task loads across satellites. This asymmetric workload leads to inefficient resource utilization and …

cn (code pays fourni par la source)

Hybrid-Action Multi-Agent Reinforcement Learning for Simultaneous Task Offloading and Resource Allocation in Space Computing Power Networks

Junyu Lai, Hao Lang, Huasuo Liu, Di Wu et autres

cn (code pays fourni par la source)

QoE-Oriented Streaming Power Allocation in LEO Satellite Networks

Huazhi Feng, Feng Wang, Junyu Lai, Qi Wu et autres

Low Earth Orbit (LEO) satellite networks are expected to support global multimedia services in future sixth-generation (6G) systems, where streaming media constitutes a major traffic type with stringent Quality of Experience (QoE) requirements. However, existing satellite resource allocation schemes mainly optimize Quality …

cn (code pays fourni par la source)

Edge Intelligence from Space: A Survey and Future Challenges

Huashuo Liu, Sheng Sun, Yali Chen, Yuwei Wang et autres

In the NewSpace era, the rapid development of satellite constellations is advancing toward seamless global coverage. In terrestrial ICT services, edge intelligence (EI) enables efficient data processing by pushing intelligent models to the network edge. Researchers are increasingly focusing on the mutual …

cn (code pays fourni par la source)

LLM En-powered UAV Networks Control for Efficient Wireless Communication

Yang Zhang, Haibo Mei, Junyu Lai, Shuang Du

Currently, Large Language Model (LLM), as a typical Generative AI (GAI) implementation, has been deployed to lead to artificial general intelligence for numerous applications. Numbers of LLM releases, like ChatGPT, Llama, Gemini, ChatGLM, can online interact with users to give instructions against …

cn (code pays fourni par la source)

Scalable Kernel Integrity Monitor

Junyu Lai, Jui-An Chang, Yu‐Sung Wu, Ying–Dar Lin et autres

The use of virtual machines in cloud data centers has become the de facto architecture for hosting AI applications. As these applications increasingly process vast amounts of data and interact with the physical world, ensuring the integrity of their underlying computing environment, …

tw, jp (code pays fourni par la source)

Computing Offloading with Edge Layering Mechanism for Mobile Ad-Hoc Networks

Yujie Liu, Junyu Lai, Hui Li, Yuhang Chen

Mobile ad-hoc networks (MANETs) encounter significant challenges in supporting efficient edge computing due to their dynamic topologies, resource limitations, and constrained bandwidth. To address these challenges, this paper proposes a novel edge layering mechanism (ELM) combined with a deep reinforcement learning (DRL)-based …

cn (code pays fourni par la source)

DFDNet: Disentangling and Filtering Dynamics for Enhanced Video Prediction

Lianqiang Gan, Junyu Lai, J. W. Ju, Yi Bin

Videos inherently contain complex temporal dynamics across various spatial directions, often entangled in ways that obscure effective dynamic extraction. Previous studies typically process video spatiotemporal features without disentangling, which hampers their ability to extract dynamic information. Additionally, the extraction of dynamics is …

cn (code pays fourni par la source)

Motion Direction Awareness: A Biomimetic Dynamic Capture Mechanism for Video Prediction

Lianqiang Gan, Junyu Lai, Junhong Zhu, Huashuo Liu et autres

Video prediction is an important yet challenging task that generates future frames based on previous observations. Despite recent progress, existing methods still suffer from motion blur, due to weak motion perception capabilities leading to uncertainty in motion direction. To address this, we …

cn (code pays fourni par la source)

Joint Computation Offloading and Resource Allocation for LEO Satellite Networks Using Hierarchical Multi-Agent Reinforcement Learning

Junyu Lai, Huashuo Liu, Guoyao Xu, Weiwei Jiang et autres

The integration of edge computing with LEO satellite broadband networks (LSBNs) offers a transformative potential, yet remains underexplored in the optimization of joint computation offloading and resource allocation (JCORA). This problem is compounded by the issues of load imbalance and hybrid action …

cn (code pays fourni par la source)

Executing Arithmetic: Fine-Tuning Large Language Models as Turing Machines

Junyu Lai, Jiahe Xu, Yao Yang, Yunpeng Huang et autres

Large Language Models (LLMs) have demonstrated remarkable capabilities across a wide range of natural language processing and reasoning tasks. However, their performance in the foundational domain of arithmetic remains unsatisfactory. When dealing with arithmetic tasks, LLMs often memorize specific examples rather than …