Ziyang Zeng

Ziyang Zeng

Auto-generated profile, based on publicly available information

Ziyang Zeng

New York University, USA

Research Focus & Metrics

Academic Career Length

2Years (2025-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
29 December 2025 in Preprints.org
Verified
Preprint
Open access
PDF file
JCJinyu Chen
Jinyu Chen
JYJixiao Yang
Jixiao Yang
ZZZiyang Zeng
Ziyang Zeng
ZHZixiao Huang
Zixiao Huang
JLJinming Li
Jinming Li
YWYutong Wang
Yutong Wang
Large Language Model (LLM)-based multi-agent systems have demonstrated remarkable capabilities across di- verse applications, yet they face critical security challenges in- cluding backdoor attacks, prompt injection, and privacy leakage. Existing defense mechanisms typically address single threat vec- tors, lacking a unified governance architecture for comprehensive security. We propose SecureGov-Agent, a governance-centric multi-agent framework that introduces a dedicated Governance Agent responsible for monitoring inter-agent communications, auditing tool invocations, and enforcing security policies. Our framework incorporates a multi-perspective risk scoring mech- anism that evaluates content risk, privacy risk, and behavioral anomalies to dynamically assess each agent’s trustworthiness. We further enhance robustness through adversarial training on syn- thesized attack scenarios. Extensive experiments across medical consultation, financial advisory, and document processing scenar- ios demonstrate that SecureGov-Agent achieves a balanced trade- off between security, privacy, and efficiency: reducing attack success rates by 73.2% compared to unprotected systems and privacy leakage rates by 81.4%, while maintaining 89.7% task completion rate with only 15.3% latency overhead. Notably, our framework excels in privacy protection (6.8% leakage rate) and maintains practical efficiency, offering a comprehensive solution for privacy-sensitive multi-agent deployments. Our framework provides a reproducible benchmark for multi-agent security research and offers practical deployment guidelines for privacy- sensitive applications.

17 December 2025 in Frontiers in Artificial Intelligence Research
Verified
Journal Article
Open access
PDF file
ZZZiyang Zeng
Ziyang Zeng
GDGuanyu Ding
Guanyu Ding
SYShiyu Yang
Shiyu Yang
The rapid expansion of digital creator economies has necessitated sophisticated engineering approaches to build platforms capable of supporting billion-user scales while maintaining robust monetization and content protection mechanisms. This review examines the full-stack engineering patterns employed in developing large-scale creator platforms, focusing on the technical architectures that enable creator monetization and content protection systems. Modern platforms face unprecedented challenges in balancing user experience, creator compensation, and intellectual property protection while managing massive concurrent user loads. This paper synthesizes current research on microservices architecture (MSA), distributed systems, blockchain technology, machine learning (ML), and cloud-native infrastructure to present comprehensive engineering patterns for building scalable creator platforms. The review analyzes payment processing systems, content delivery networks (CDN), digital rights management (DRM), and fraud detection mechanisms that constitute the technical foundation of billion-user products. Through examination of contemporary platforms and emerging technologies, this paper identifies critical design patterns including event-driven architecture (EDA), eventual consistency models, horizontal scaling strategies, and zero-trust security frameworks. The findings reveal that successful billion-user creator platforms require careful orchestration of backend infrastructure, real-time data processing, intelligent caching strategies, and multi-layered security approaches. This review contributes to the understanding of engineering tradeoffs in building creator economy platforms and provides insights into future technological directions for supporting the next generation of digital content creation and distribution systems.