arXiv subjects

Cyril Leung

Publications and source records attributed to Cyril Leung.

At least 19 recordsLinked to original sources

arXiv2026

QoS-Aware Token Scheduling and Private Data Valuation for Multi-Modal Agentic Networks

In agentic systems, human-generated data records anchor the value of AI services. Yet cloud compute pipelines centralize processing on remote servers. Data centralization reduces personal data sovereignty and may potentially degrade the quality of service (QoS). Meanwhile, user contributions are diverse in quantity and quality: decentralized records can be biased, noisy, and heterogeneously distributed. To address the data challenge, we study fair token allocation and private data valuation for decentralized and resource-constrained agentic systems. Our approach embeds multi-modal representations in a shared semantic space and releases differentially private (DP) prototypes to preserve utility while reducing semantic leakage. With the DP guarantee, we design a fair token allocation scheme that rewards effective contributions and remains robust to data heterogeneity and AI resource scarcity. Extensive simulations demonstrate improved contribution-based fairness and QoS compared to standard benchmarks. The improved resistance to image reconstruction attacks indicates enhanced privacy for multi-modal personal data.

cs.AI↗

arXiv2026

VENOMREC: Cross-Modal Interactive Poisoning for Targeted Promotion in Multimodal LLM Recommender Systems

Multimodal large language models (MLLMs) are pushing recommender systems (RecSys) toward content-grounded retrieval and ranking via cross-modal fusion. We find that while cross-modal consensus often mitigates conventional poisoning that manipulates interaction logs or perturbs a single modality, it also introduces a new attack surface where synchronised multimodal poisoning can reliably steer fused representations along stable semantic directions during fine-tuning. To characterise this threat, we formalise cross-modal interactive poisoning and propose VENOMREC, which performs Exposure Alignment to identify high-exposure regions in the joint embedding space and Cross-modal Interactive Perturbation to craft attention-guided coupled token--patch edits. Experiments on four real-world multimodal datasets demonstrate that VENOMREC consistently outperforms strong baselines, achieving 0.73 mean ER@20 and improving over the strongest baseline by +0.52 absolute ER points on average, while maintaining comparable recommendation utility. Code is available at https://github.com/GuoweiGuan666/VenomRec.

cs.CR↗

arXiv2025

An Efficient Interaction Human-AI Synergy System Bridging Visual Awareness and Large Language Model for Intensive Care Units

Intensive Care Units (ICUs) are critical environments characterized by high-stakes monitoring and complex data management. However, current practices often rely on manual data transcription and fragmented information systems, introducing potential risks to patient safety and operational efficiency. To address these issues, we propose a human-AI synergy system based on a cloud-edge-end architecture, which integrates visual-aware data extraction and semantic interaction mechanisms. Specifically, a visual-aware edge module non-invasively captures real-time physiological data from bedside monitors, reducing manual entry errors. To improve accessibility to fragmented data sources, a semantic interaction module, powered by a Large Language Model (LLM), enables physicians to perform efficient and intuitive voice-based queries over structured patient data. The hierarchical cloud-edge-end deployment ensures low-latency communication and scalable system performance. Our system reduces the cognitive burden on ICU nurses and physicians and demonstrates promising potential for broader applications in intelligent healthcare systems.

cs.HC↗

arXiv2024

Update Rate, Accuracy, and Age of Information in a Wireless Sensor Network

Age of Information (AoI), namely the time that has elapsed since the most recently delivered packet was generated, is receiving increasing attention with the emergence of many real-time applications that rely on the exchange of time-sensitive information. AoI captures the freshness of the information from the perspective of the destination. The term "accuracy of information" is used to assess how close the estimate at the destination is to the parameter value measured by the sensor. In this paper, the mean square error (MSE) is used to evaluate the accuracy of information. We focus on a single sensor that monitors a time-sensitive physical process, which is modelled as a random walk. Whenever the state of the random walk changes by more than a specified threshold, the sensor generates a status update packet and transmits it to the destination. When no update packet is received, the destination assumes that the state of the process has not changed. We study the problem of finding the minimum update rate under AoI and accuracy of information constraints. More specifically, we derive analytical expressions for the update rate, the AoI, and the MSE.

cs.IT↗

arXiv2023

Can ChatGPT Assess Human Personalities? A General Evaluation Framework

Large Language Models (LLMs) especially ChatGPT have produced impressive results in various areas, but their potential human-like psychology is still largely unexplored. Existing works study the virtual personalities of LLMs but rarely explore the possibility of analyzing human personalities via LLMs. This paper presents a generic evaluation framework for LLMs to assess human personalities based on Myers Briggs Type Indicator (MBTI) tests. Specifically, we first devise unbiased prompts by randomly permuting options in MBTI questions and adopt the average testing result to encourage more impartial answer generation. Then, we propose to replace the subject in question statements to enable flexible queries and assessments on different subjects from LLMs. Finally, we re-formulate the question instructions in a manner of correctness evaluation to facilitate LLMs to generate clearer responses. The proposed framework enables LLMs to flexibly assess personalities of different groups of people. We further propose three evaluation metrics to measure the consistency, robustness, and fairness of assessment results from state-of-the-art LLMs including ChatGPT and GPT-4. Our experiments reveal ChatGPT's ability to assess human personalities, and the average results demonstrate that it can achieve more consistent and fairer assessments in spite of lower robustness against prompt biases compared with InstructGPT.

cs.CL↗

arXiv2023

A Survey of Computer Vision Technologies In Urban and Controlled-environment Agriculture

In the evolution of agriculture to its next stage, Agriculture 5.0, artificial intelligence will play a central role. Controlled-environment agriculture, or CEA, is a special form of urban and suburban agricultural practice that offers numerous economic, environmental, and social benefits, including shorter transportation routes to population centers, reduced environmental impact, and increased productivity. Due to its ability to control environmental factors, CEA couples well with computer vision (CV) in the adoption of real-time monitoring of the plant conditions and autonomous cultivation and harvesting. The objective of this paper is to familiarize CV researchers with agricultural applications and agricultural practitioners with the solutions offered by CV. We identify five major CV applications in CEA, analyze their requirements and motivation, and survey the state of the art as reflected in 68 technical papers using deep learning methods. In addition, we discuss five key subareas of computer vision and how they related to these CEA problems, as well as eleven vision-based CEA datasets. We hope the survey will help researchers quickly gain a bird-eye view of the striving research area and will spark inspiration for new research and development.

cs.CV↗

arXiv2023

Hierarchical Skeleton Meta-Prototype Contrastive Learning with Hard Skeleton Mining for Unsupervised Person Re-Identification

With rapid advancements in depth sensors and deep learning, skeleton-based person re-identification (re-ID) models have recently achieved remarkable progress with many advantages. Most existing solutions learn single-level skeleton features from body joints with the assumption of equal skeleton importance, while they typically lack the ability to exploit more informative skeleton features from various levels such as limb level with more global body patterns. The label dependency of these methods also limits their flexibility in learning more general skeleton representations. This paper proposes a generic unsupervised Hierarchical skeleton Meta-Prototype Contrastive learning (Hi-MPC) approach with Hard Skeleton Mining (HSM) for person re-ID with unlabeled 3D skeletons. Firstly, we construct hierarchical representations of skeletons to model coarse-to-fine body and motion features from the levels of body joints, components, and limbs. Then a hierarchical meta-prototype contrastive learning model is proposed to cluster and contrast the most typical skeleton features ("prototypes") from different-level skeletons. By converting original prototypes into meta-prototypes with multiple homogeneous transformations, we induce the model to learn the inherent consistency of prototypes to capture more effective skeleton features for person re-ID. Furthermore, we devise a hard skeleton mining mechanism to adaptively infer the informative importance of each skeleton, so as to focus on harder skeletons to learn more discriminative skeleton representations. Extensive evaluations on five datasets demonstrate that our approach outperforms a wide variety of state-of-the-art skeleton-based methods. We further show the general applicability of our method to cross-view person re-ID and RGB-based scenarios with estimated skeletons.

cs.CV↗

arXiv2023

Towards Fairness-Aware Federated Learning

Recent advances in Federated Learning (FL) have brought large-scale collaborative machine learning opportunities for massively distributed clients with performance and data privacy guarantees. However, most current works focus on the interest of the central controller in FL,and overlook the interests of the FL clients. This may result in unfair treatment of clients that discourages them from actively participating in the learning process and damages the sustainability of the FL ecosystem. Therefore, the topic of ensuring fairness in FL is attracting a great deal of research interest. In recent years, diverse Fairness-Aware FL (FAFL) approaches have been proposed in an effort to achieve fairness in FL from different perspectives. However, there is no comprehensive survey that helps readers gain insight into this interdisciplinary field. This paper aims to provide such a survey. By examining the fundamental and simplifying assumptions, as well as the notions of fairness adopted by existing literature in this field, we propose a taxonomy of FAFL approaches covering major steps in FL, including client selection, optimization, contribution evaluation and incentive distribution. In addition, we discuss the main metrics for experimentally evaluating the performance of FAFL approaches, and suggest promising future research directions towards FAFL.

cs.LG↗

arXiv2022

Inductive Graph Transformer for Delivery Time Estimation

Providing accurate estimated time of package delivery on users' purchasing pages for e-commerce platforms is of great importance to their purchasing decisions and post-purchase experiences. Although this problem shares some common issues with the conventional estimated time of arrival (ETA), it is more challenging with the following aspects: 1) Inductive inference. Models are required to predict ETA for orders with unseen retailers and addresses; 2) High-order interaction of order semantic information. Apart from the spatio-temporal features, the estimated time also varies greatly with other factors, such as the packaging efficiency of retailers, as well as the high-order interaction of these factors. In this paper, we propose an inductive graph transformer (IGT) that leverages raw feature information and structural graph data to estimate package delivery time. Different from previous graph transformer architectures, IGT adopts a decoupled pipeline and trains transformer as a regression function that can capture the multiplex information from both raw feature and dense embeddings encoded by a graph neural network (GNN). In addition, we further simplify the GNN structure by removing its non-linear activation and the learnable linear transformation matrix. The reduced parameter search space and linear information propagation in the simplified GNN enable the IGT to be applied in large-scale industrial scenarios. Experiments on real-world logistics datasets show that our proposed model can significantly outperform the state-of-the-art methods on estimation of delivery time. The source code is available at: https://github.com/enoche/IGT-WSDM23.

cs.LG↗

arXiv2022

Exploring Representation-Level Augmentation for Code Search

Code search, which aims at retrieving the most relevant code fragment for a given natural language query, is a common activity in software development practice. Recently, contrastive learning is widely used in code search research, where many data augmentation approaches for source code (e.g., semantic-preserving program transformation) are proposed to learn better representations. However, these augmentations are at the raw-data level, which requires additional code analysis in the preprocessing stage and additional training costs in the training stage. In this paper, we explore augmentation methods that augment data (both code and query) at representation level which does not require additional data processing and training, and based on this we propose a general format of representation-level augmentation that unifies existing methods. Then, we propose three new augmentation methods (linear extrapolation, binary interpolation, and Gaussian scaling) based on the general format. Furthermore, we theoretically analyze the advantages of the proposed augmentation methods over traditional contrastive learning methods on code search. We experimentally evaluate the proposed representation-level augmentation methods with state-of-the-art code search models on a large-scale public dataset consisting of six programming languages. The experimental results show that our approach can consistently boost the performance of the studied code search models. Our source code is available at https://github.com/Alex-HaochenLi/RACS.

cs.SE↗

arXiv2022

A Dynamic Hierarchical Framework for IoT-assisted Metaverse Synchronization

Metaverse has recently attracted much attention from both academia and industry. Virtual services, ranging from virtual driver training to online route optimization for smart goods delivery, are emerging in the Metaverse. To make the human experience of virtual life more real, digital twins (DTs), namely digital replicas of physical objects, are key enablers. However, DT status may not always accurately reflect that of its real-world twin because the latter may be subject to changes with time. As such, it is necessary to synchronize a DT with its physical counterpart to ensure that its status is accurate for virtual businesses in the Metaverse. In this paper, we propose a dynamic hierarchical framework in which a group of IoT devices is incentivized to sense and collect physical objects' status information collectively so as to assists virtual service providers (VSPs) in synchronizing DTs. Based on the collected sensing data and the value decay rate of the DTs, the VSPs can determine synchronization intensities to maximize their payoffs. In our proposed dynamic hierarchical framework, the lower-level evolutionary game captures the VSPs selection by the IoT device population, and the upper-level differential game captures the VSPs payoffs, which are affected by the synchronization strategy, IoT devices selections, and the DTs value status, given VSPs are simultaneous decision makers. We further consider the case in which some VSPs are first movers and extend it as a Stackelberg differential game. We theoretically and experimentally show that the equilibrium to the lower-level game exists and is evolutionarily robust, and provide a sensitivity analysis with respect to various system parameters. Experiments show that the proposed dynamic hierarchical game outperform the baseline.

cs.GT↗

arXiv2021

A Dynamic Resource Allocation Framework for Synchronizing Metaverse with IoT Service and Data

Spurred by the severe restrictions on mobility due to the COVID-19 pandemic, there is currently intense interest in developing the Metaverse, to offer virtual services/business online. A key enabler of such virtual service is the digital twin, i.e., a digital replication of real-world entities in the Metaverse, e.g., city twin, avatars, etc. The real-world data collected by IoT devices and sensors are key for synchronizing the two worlds. In this paper, we consider the scenario in which a group of IoT devices are employed by the Metaverse platform to collect such data on behalf of virtual service providers (VSPs). Device owners, who are self-interested, dynamically select a VSP to maximize rewards. We adopt hybrid evolutionary dynamics, in which heterogeneous device owner populations can employ different revision protocols to update their strategies. Extensive simulations demonstrate that a hybrid protocol can lead to evolutionary stable states.

cs.GT↗

arXiv2021

Dynamics in Coded Edge Computing for IoT: A Fractional Evolutionary Game Approach

Recently, coded distributed computing (CDC), with advantages in intensive computation and reduced latency, has attracted a lot of research interest for edge computing, in particular, IoT applications, including IoT data pre-processing and data analytics. Nevertheless, it can be challenging for edge infrastructure providers (EIPs) with limited edge resources to support IoT applications performed in a CDC approach in edge networks, given the additional computational resources required by CDC. In this paper, we propose coded edge federation, in which different EIPs collaboratively provide edge resources for CDC tasks. To study the Nash equilibrium, when no EIP has an incentive to unilaterally alter its decision on edge resource allocation, we model the coded edge federation based on evolutionary game theory. Since the replicator dynamics of the classical evolutionary game are unable to model economic-aware EIPs which memorize past decisions and utilities, we propose fractional replicator dynamics with a power-law fading memory via Caputo fractional derivatives. The proposed dynamics allow us to study a broad spectrum of EIP dynamic behaviors, such as EIP sensitivity and aggressiveness in strategy adaptation, which classical replicator dynamics cannot capture. Theoretical analysis and extensive numerical results justify the existence, uniqueness, and stability of the equilibrium in the fractional evolutionary game. The influence of the content and the length of the memory on the rate of convergence is also investigated.

cs.DC↗

arXiv2021

A Contract Theory based Incentive Mechanism for Federated Learning

Federated learning (FL) serves as a data privacy-preserved machine learning paradigm, and realizes the collaborative model trained by distributed clients. To accomplish an FL task, the task publisher needs to pay financial incentives to the FL server and FL server offloads the task to the contributing FL clients. It is challenging to design proper incentives for the FL clients due to the fact that the task is privately trained by the clients. This paper aims to propose a contract theory based FL task training model towards minimizing incentive budget subject to clients being individually rational (IR) and incentive compatible (IC) in each FL training round. We design a two-dimensional contract model by formally defining two private types of clients, namely data quality and computation effort. To effectively aggregate the trained models, a contract-based aggregator is proposed. We analyze the feasible and optimal contract solutions to the proposed contract model. %Experimental results demonstrate that the proposed framework and contract model can effective improve the generation accuracy of FL tasks. Experimental results show that the generalization accuracy of the FL tasks can be improved by the proposed incentive mechanism where contract-based aggregation is applied.

cs.DC↗

arXiv2021

A Game-theoretic Approach Towards Collaborative Coded Computation Offloading

Coded distributed computing (CDC) has emerged as a promising approach because it enables computation tasks to be carried out in a distributed manner while mitigating straggler effects, which often account for the long overall completion times. Specifically, by using polynomial codes, computed results from only a subset of edge servers can be used to reconstruct the final result. However, incentive issues have not been studied systematically for the edge servers to complete the CDC tasks. In this paper, we propose a tractable two-level game-theoretic approach to incentivize the edge servers to complete the CDC tasks. Specifically, in the lower level, a hedonic coalition formation game is formulated where the edge servers share their resources within their coalitions. By forming coalitions, the edge servers have more Central Processing Unit (CPU) power to complete the computation tasks. In the upper level, given the CPU power of the coalitions of edge servers, an all-pay auction is designed to incentivize the edge servers to participate in the CDC tasks. In the all-pay auction, the bids of the edge servers are represented by the allocation of their CPU power to the CDC tasks. The all-pay auction is designed to maximize the utility of the cloud server by determining the allocation of rewards to the winners. Simulation results show that the edge servers are incentivized to allocate more CPU power when multiple rewards are offered, i.e., there are multiple winners, instead of rewarding only the edge server with the largest CPU power allocation. Besides, the utility of the cloud server is maximized when it offers multiple homogeneous rewards, instead of heterogeneous rewards.

cs.GT↗

arXiv2020

Collaborative Coded Computation Offloading: An All-pay Auction Approach

As the amount of data collected for crowdsensing applications increases rapidly due to improved sensing capabilities and the increasing number of Internet of Things (IoT) devices, the cloud server is no longer able to handle the large-scale datasets individually. Given the improved computational capabilities of the edge devices, coded distributed computing has become a promising approach given that it allows computation tasks to be carried out in a distributed manner while mitigating straggler effects, which often account for the long overall completion times. Specifically, by using polynomial codes, computed results from only a subset of devices are needed to reconstruct the final result. However, there is no incentive for the edge devices to complete the computation tasks. In this paper, we present an all-pay auction to incentivize the edge devices to participate in the coded computation tasks. In this auction, the bids of the edge devices are represented by the allocation of their Central Processing Unit (CPU) power to the computation tasks. All edge devices submit their bids regardless of whether they win or lose in the auction. The all-pay auction is designed to maximize the utility of the cloud server by determining the reward allocation to the winners. Simulation results show that the edge devices are incentivized to allocate more CPU power when multiple rewards are offered instead of a single reward.

cs.DC↗

arXiv2020

Optimal Transaction Queue Waiting in Blockchain Mining

Blockchain systems are being used in a wide range of application domains. They can support trusted transactions in time critical applications. In this paper, we study how miners should pick up transactions from a transaction pool so as to minimize the average waiting time per transaction. We derive an expression for the average transaction waiting time of the proposed mining scheme and determine the optimum decision rule. Numerical results show that the average waiting time per transaction can be reduced by about 10% compared to the traditional no-wait scheme in which miners immediately start the next mining round using all transactions waiting in the pool.

cs.PF↗

arXiv2020

A Survey of Coded Distributed Computing

Distributed computing has become a common approach for large-scale computation of tasks due to benefits such as high reliability, scalability, computation speed, and costeffectiveness. However, distributed computing faces critical issues related to communication load and straggler effects. In particular, computing nodes need to exchange intermediate results with each other in order to calculate the final result, and this significantly increases communication overheads. Furthermore, a distributed computing network may include straggling nodes that run intermittently slower. This results in a longer overall time needed to execute the computation tasks, thereby limiting the performance of distributed computing. To address these issues, coded distributed computing (CDC), i.e., a combination of coding theoretic techniques and distributed computing, has been recently proposed as a promising solution. Coding theoretic techniques have proved effective in WiFi and cellular systems to deal with channel noise. Therefore, CDC may significantly reduce communication load, alleviate the effects of stragglers, provide fault-tolerance, privacy and security. In this survey, we first introduce the fundamentals of CDC, followed by basic CDC schemes. Then, we review and analyze a number of CDC approaches proposed to reduce the communication costs, mitigate the straggler effects, and guarantee privacy and security. Furthermore, we present and discuss applications of CDC in modern computer networks. Finally, we highlight important challenges and promising research directions related to CDC

cs.DC↗