arXiv subjects

Sheng Wei

Publications and source records attributed to Sheng Wei.

5 recordsLinked to original sources

arXiv2026

Causal Discovery as Dialectical Aggregation: A Quantitative Argumentation Framework

Constraint-based causal discovery is brittle in finite-sample regimes because erroneous conditional-independence (CI) decisions can cascade into substantial structural errors. We propose Quantitative Argumentation for Causal Discovery (QACD), a semantics-driven framework that represents CI outcomes as graded, defeasible arguments rather than irreversible constraints. QACD maps statistical test outcomes to argument strengths and aggregates conflicting evidence through connectivity-mediated witness propagation, producing a fixed-point acceptability labeling over candidate adjacencies. Experiments on standard benchmark Bayesian networks suggest that QACD improves structural coherence and interventional reliability in several noisy or inconsistent CI regimes, while remaining competitive with classical constraint-based, hybrid, and prior argumentation-based baselines.

cs.AI↗

arXiv2025

A Distributed Framework for Privacy-Enhanced Vision Transformers on the Edge

Nowadays, visual intelligence tools have become ubiquitous, offering all kinds of convenience and possibilities. However, these tools have high computational requirements that exceed the capabilities of resource-constrained mobile and wearable devices. While offloading visual data to the cloud is a common solution, it introduces significant privacy vulnerabilities during transmission and server-side computation. To address this, we propose a novel distributed, hierarchical offloading framework for Vision Transformers (ViTs) that addresses these privacy challenges by design. Our approach uses a local trusted edge device, such as a mobile phone or an Nvidia Jetson, as the edge orchestrator. This orchestrator partitions the user's visual data into smaller portions and distributes them across multiple independent cloud servers. By design, no single external server possesses the complete image, preventing comprehensive data reconstruction. The final data merging and aggregation computation occurs exclusively on the user's trusted edge device. We apply our framework to the Segment Anything Model (SAM) as a practical case study, which demonstrates that our method substantially enhances content privacy over traditional cloud-based approaches. Evaluations show our framework maintains near-baseline segmentation performance while substantially reducing the risk of content reconstruction and user data exposure. Our framework provides a scalable, privacy-preserving solution for vision tasks in the edge-cloud continuum.

cs.DC↗

arXiv2025

Human-AI Collaboration in Cloud Security: Cognitive Hierarchy-Driven Deep Reinforcement Learning

Given the complexity of multi-tenant cloud environments and the growing need for real-time threat mitigation, Security Operations Centers (SOCs) must adopt AI-driven adaptive defense mechanisms to counter Advanced Persistent Threats (APTs). However, SOC analysts face challenges in handling adaptive adversarial tactics, requiring intelligent decision-support frameworks. We propose a Cognitive Hierarchy Theory-driven Deep Q-Network (CHT-DQN) framework that models interactive decision-making between SOC analysts and AI-driven APT bots. The SOC analyst (defender) operates at cognitive level-1, anticipating attacker strategies, while the APT bot (attacker) follows a level-0 policy. By incorporating CHT into DQN, our framework enhances adaptive SOC defense using Attack Graph (AG)-based reinforcement learning. Simulation experiments across varying AG complexities show that CHT-DQN consistently achieves higher data protection and lower action discrepancies compared to standard DQN. A theoretical lower bound further confirms its superiority as AG complexity increases. A human-in-the-loop (HITL) evaluation on Amazon Mechanical Turk (MTurk) reveals that SOC analysts using CHT-DQN-derived transition probabilities align more closely with adaptive attackers, leading to better defense outcomes. Moreover, human behavior aligns with Prospect Theory (PT) and Cumulative Prospect Theory (CPT): participants are less likely to reselect failed actions and more likely to persist with successful ones. This asymmetry reflects amplified loss sensitivity and biased probability weighting -- underestimating gains after failure and overestimating continued success. Our findings highlight the potential of integrating cognitive models into deep reinforcement learning to improve real-time SOC decision-making for cloud security.

cs.CR↗

arXiv2024

Privacy-Preserving Multimedia Mobile Cloud Computing Using Protective Perturbation

Mobile cloud computing has been adopted in many multimedia applications, where the resource-constrained mobile device sends multimedia data (e.g., images) to remote cloud servers to request computation-intensive multimedia services (e.g., image recognition). While significantly improving the performance of the mobile applications, the cloud-based mechanism often causes privacy concerns as the multimedia data and services are offloaded from the trusted user device to untrusted cloud servers. Several recent studies have proposed perturbation-based privacy preserving mechanisms, which obfuscate the offloaded multimedia data to eliminate privacy exposures without affecting the functionality of the remote multimedia services. However, the existing privacy protection approaches require the deployment of computation-intensive perturbation generation on the resource-constrained mobile devices. Also, the obfuscated images are typically not compliant with the standard image compression algorithms and suffer from significant bandwidth consumption. In this paper, we develop a novel privacy-preserving multimedia mobile cloud computing framework, namely $PMC^2$, to address the resource and bandwidth challenges. $PMC^2$ employs secure confidential computing in the cloud to deploy the perturbation generator, which addresses the resource challenge while maintaining the privacy. Furthermore, we develop a neural compressor specifically trained to compress the perturbed images in order to address the bandwidth challenge. We implement $PMC^2$ in an end-to-end mobile cloud computing system, based on which our evaluations demonstrate superior latency, power efficiency, and bandwidth consumption achieved by $PMC^2$ while maintaining high accuracy in the target multimedia service.

cs.MM↗

arXiv2022

Privacy-preserving Reflection Rendering for Augmented Reality

Many augmented reality (AR) applications rely on omnidirectional environment lighting to render photorealistic virtual objects. When the virtual objects consist of reflective materials, such as a metallic sphere, the required lighting information to render such objects can consist of privacy-sensitive information that is outside the current camera view. In this paper, we show, for the first time, that accuracy-driven multi-view environment lighting can reveal out-of-camera scene information and compromise privacy. We present a simple yet effective privacy attack that extracts sensitive scene information such as human face and text information from the rendered objects, under a number of application scenarios. To defend against such attacks, we develop a novel $IPC^{2}S$ defense and a conditional $R^2$ defense. Our $IPC^{2}S$ defense, used in conjunction with a generic lighting reconstruction method, preserves the scene geometry while obfuscating the privacy-sensitive information. As a proof-of-concept, we leverage existing OCR and face detection models to identify text and human faces from past camera observations and blur the color pixels associated with detected regions. We evaluate the visual quality impact of our defense by comparing rendered virtual objects to ones rendered with a generic multi-lighting reconstruction technique, ARKit, and $R^2$ defense. Our visual and quantitative results demonstrate that our defense leads to structurally similar reflections with up to 0.98 SSIM score across a variety of rendering scenarios while preserving sensitive information by reducing the automatic extraction success rate to at most 8.8%.

cs.MM↗