Rui-Xiao Zhang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Rui-Xiao Zhang has a total of 58 co-authors. Here are the top co-authors by number of shared publications:
- Tianchi Huang
Tsinghua University, Beijing, China
7 shared publications - Lifeng Sun
Tsinghua University, Beijing, China
6 shared publications - Klara Nahrstedt
University of Illinois Urbana-Champaign, Urbana, USA
5 shared publications - Shu Shi
Bytedance, San Diego, USA
5 shared publications - Haiping Wang
Bytedance, Beijing, China
4 shared publications - Chenglei Wu
Kuaishou Technology, Hangzhou, China
3 shared publications
Latest Publications
- Open Access
- 2 September 2026
- Published by Association for Computing Machinery (ACM) in Proceedings of the ACM on Networking
- Vol. 4 (CoNEXT3) , 1-25
Generative Active Adaptation for Drifting and Imbalanced Network Intrusion Detection
Machine learning–based network intrusion detection systems (NIDS) monitor network traffic to identify malicious activity. While NIDS perform well in controlled environments, their effectiveness deteriorates in real-world deployments due to concept drift. As the underlying distribution evolves and new attacks emerge, benign traffic nevertheless remains dominant, so rare and emerging attacks are sparsely observed. This severe class imbalance makes it difficult to identify which data should be prioritized for model adaptation and often leads to high labeling cost, as identifying the few informative attacks requires extensive manual inspection. We present NetGuard, a generative active adaptation framework that enables robust and label-efficient adaptation of NIDS under dynamic network conditions. NetGuard combines density-aware active sampling to identify informative drifted traffic for annotation with deep generative modeling to synthesize diverse, minority-class samples with high-fidelity to expand coverage of rare attacks. By jointly selecting what to label and how to augment scarce data, NetGuard mitigates performance degradation caused by both distribution shift and long-tailed attack distributions. We carry out a comprehensive evaluation of NetGuard across intrusion detection datasets including a large-scale real-world ISP dataset. On average, NetGuard improves overall detection performance by 2×, while achieving 8–15× higher F1 and 4–7× lower false-negative rates for rare and emerging attacks compared to state-of-the-art sample selection baselines, using only 1% labeled data. NetGuard enhances rare attack detection while reducing labeling costs, rendering it scalable and practical for intrusion detection.
- 2 September 2026
- Published by Association for Computing Machinery (ACM) in Proceedings of the ACM on Networking
- Vol. 4 (CoNEXT3) , 1-25
Open AccessGenerative Active Adaptation for Drifting and Imbalanced Network Intrusion Detection
Machine learning–based network intrusion detection systems (NIDS) monitor network traffic to identify malicious activity. While NIDS perform well in controlled environments, their effectiveness deteriorates in real-world deployments due to concept drift. As the underlying distribution evolves and new attacks emerge, benign traffic nevertheless remains dominant, so rare and emerging attacks are sparsely observed. This severe class imbalance makes it difficult to identify which data should be prioritized for model adaptation and often leads to high labeling cost, as identifying the few informative attacks requires extensive manual inspection. We present NetGuard, a generative active adaptation framework that enables robust and label-efficient adaptation of NIDS under dynamic network conditions. NetGuard combines density-aware active sampling to identify informative drifted traffic for annotation with deep generative modeling to synthesize diverse, minority-class samples with high-fidelity to expand coverage of rare attacks. By jointly selecting what to label and how to augment scarce data, NetGuard mitigates performance degradation caused by both distribution shift and long-tailed attack distributions. We carry out a comprehensive evaluation of NetGuard across intrusion detection datasets including a large-scale real-world ISP dataset. On average, NetGuard improves overall detection performance by 2×, while achieving 8–15× higher F1 and 4–7× lower false-negative rates for rare and emerging attacks compared to state-of-the-art sample selection baselines, using only 1% labeled data. NetGuard enhances rare attack detection while reducing labeling costs, rendering it scalable and practical for intrusion detection.
- Open Access
- 22 May 2026
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Computer Systems
- Vol. 44 (3) , 1-29
Managing Edge Resources at Scale: A Peer-to-Peer CDN for On-Demand Video Streaming
The rapid proliferation of video services has created a strong demand for efficient content delivery. Traditional solutions primarily rely on Content Delivery Networks (CDNs), which, while effective, impose substantial bandwidth costs on video providers. To address this challenge, cost-efficient edge resources have emerged as a promising alternative. However, their heterogeneous hardware and inherently limited performance pose significant challenges for practical deployment. In this paper, we present how ByteDance explores the use of these cost-efficient but less performant resources. Specifically, we first present an extensive overview of PCDN (Peer-to-Peer CDN), ByteDance’s alternative delivery network for CDNs. Second, as PCDN encounters speed degradation and resource imbalances after years of deployment, we further introduce PCDN + , the enhanced iteration of PCDN. Specifically, we first propose a joint control protocol. By using de-synchronized probing and proactive retransmission, our protocol significantly improves the performance degradation caused by self-competition and head-of-line blocking (HoL). We then introduce a resource management framework. By integrating a well-designed centralized/decentralized framework, we evolve previous “static” and “uncontrolled” PCDN into a “dynamic” and “controlled” system. The extensive A/B test and real-world deployment have demonstrated that PCDN + 1) provides higher service speed, 2) effectively alleviates overloading issues, 3) significantly improves the utilization of low-cost resources.
- 22 May 2026
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Computer Systems
- Vol. 44 (3) , 1-29
Open AccessManaging Edge Resources at Scale: A Peer-to-Peer CDN for On-Demand Video Streaming
The rapid proliferation of video services has created a strong demand for efficient content delivery. Traditional solutions primarily rely on Content Delivery Networks (CDNs), which, while effective, impose substantial bandwidth costs on video providers. To address this challenge, cost-efficient edge resources have emerged as a promising alternative. However, their heterogeneous hardware and inherently limited performance pose significant challenges for practical deployment. In this paper, we present how ByteDance explores the use of these cost-efficient but less performant resources. Specifically, we first present an extensive overview of PCDN (Peer-to-Peer CDN), ByteDance’s alternative delivery network for CDNs. Second, as PCDN encounters speed degradation and resource imbalances after years of deployment, we further introduce PCDN + , the enhanced iteration of PCDN. Specifically, we first propose a joint control protocol. By using de-synchronized probing and proactive retransmission, our protocol significantly improves the performance degradation caused by self-competition and head-of-line blocking (HoL). We then introduce a resource management framework. By integrating a well-designed centralized/decentralized framework, we evolve previous “static” and “uncontrolled” PCDN into a “dynamic” and “controlled” system. The extensive A/B test and real-world deployment have demonstrated that PCDN + 1) provides higher service speed, 2) effectively alleviates overloading issues, 3) significantly improves the utilization of low-cost resources.
- Open Access
- 4 April 2026
- Published by Association for Computing Machinery (ACM)
- p. 8-14
Perception-Aware Frame Display Scheduling for Low-Latency Cloud Gaming
- 4 April 2026
- Published by Association for Computing Machinery (ACM)
- p. 8-14
Open AccessPerception-Aware Frame Display Scheduling for Low-Latency Cloud Gaming
- Open Access
- 4 April 2026
- Published by Association for Computing Machinery (ACM)
- p. 25-36
T rinity : Exploiting Latency Sensitivity to Improve Quality of Experience on Cloud VR Gaming
- 4 April 2026
- Published by Association for Computing Machinery (ACM)
- p. 25-36
Open AccessT rinity : Exploiting Latency Sensitivity to Improve Quality of Experience on Cloud VR Gaming
- Open Access
- 27 October 2025
- Published by Association for Computing Machinery (ACM)
- p. 12218-12226
Generative Flow Networks for Personalized Multimedia Systems: A Case Study on Short Video Feeds
- 27 October 2025
- Published by Association for Computing Machinery (ACM)
- p. 12218-12226
Open AccessGenerative Flow Networks for Personalized Multimedia Systems: A Case Study on Short Video Feeds
- Open Access
- 27 October 2025
- Published by Association for Computing Machinery (ACM)
- p. 12045-12053
Progressive Learning with Human Feedback for Personalized Adaptive Video Streaming
- 27 October 2025
- Published by Association for Computing Machinery (ACM)
- p. 12045-12053
Open AccessProgressive Learning with Human Feedback for Personalized Adaptive Video Streaming
- Open Access
- 23 June 2025
- Published by Association for Computing Machinery (ACM)
- p. 388-401
NeRFlow: Towards Adaptive Streaming for NeRF Videos
- 23 June 2025
- Published by Association for Computing Machinery (ACM)
- p. 388-401
Open AccessNeRFlow: Towards Adaptive Streaming for NeRF Videos
- Open Access
- 4 November 2024
- Published by Association for Computing Machinery (ACM)
- p. 588-594
M agpie : Improving the Efficiency of A/B Tests for Large Scale Video-on-Demand Systems
With the exponential rise in video traffic, researchers and developers require more effective tools to validate the efficacy of designed algorithms for Video-on-Demand (VoD) system. However, traditional experimental platforms face two main challenges: a lack of realistic testing and the need for longer and significant effort. To overcome these limitations, we propose Magpie, an efficient experimental platform tailored for VoD systems. Magpie leverages a realistic operational setting, rapid testing, and high reproducibility to closely simulate online user environments without impacting production systems. Compared to conventional simulations, our evaluation demonstrates that Magpie reduces the disparity with online experiments by 85.6%. Deployed within our company-a leading video content provider in China-Magpie has efficiently validated over tens of algorithms, with 80% demonstrating enhanced performance in subsequent online tests.
- 4 November 2024
- Published by Association for Computing Machinery (ACM)
- p. 588-594
Open AccessM agpie : Improving the Efficiency of A/B Tests for Large Scale Video-on-Demand Systems
With the exponential rise in video traffic, researchers and developers require more effective tools to validate the efficacy of designed algorithms for Video-on-Demand (VoD) system. However, traditional experimental platforms face two main challenges: a lack of realistic testing and the need for longer and significant effort. To overcome these limitations, we propose Magpie, an efficient experimental platform tailored for VoD systems. Magpie leverages a realistic operational setting, rapid testing, and high reproducibility to closely simulate online user environments without impacting production systems. Compared to conventional simulations, our evaluation demonstrates that Magpie reduces the disparity with online experiments by 85.6%. Deployed within our company-a leading video content provider in China-Magpie has efficiently validated over tens of algorithms, with 80% demonstrating enhanced performance in subsequent online tests.
- Open Access
- 28 October 2024
- Published by Association for Computing Machinery (ACM)
- p. 11276-11278
Scene Graph Driven Hybrid Interactive VR Teleconferencing
We propose an interactive and intelligent hybrid teleconferencing system compatible with Virtual Reality devices. Our system understands meeting contexts and leverages user interactions to enhance better system configuration. Employing interactive scene graphs [11], the system extracts and transmits essential meeting context to users while relaying user interactions back to the streaming systems for user-involved adaptive streaming and foveated rendering. We demonstrate the system's real-time performance and compatibility with commercial VR devices such as the Meta Quest 3.
- 28 October 2024
- Published by Association for Computing Machinery (ACM)
- p. 11276-11278
Open AccessScene Graph Driven Hybrid Interactive VR Teleconferencing
We propose an interactive and intelligent hybrid teleconferencing system compatible with Virtual Reality devices. Our system understands meeting contexts and leverages user interactions to enhance better system configuration. Employing interactive scene graphs [11], the system extracts and transmits essential meeting context to users while relaying user interactions back to the streaming systems for user-involved adaptive streaming and foveated rendering. We demonstrate the system's real-time performance and compatibility with commercial VR devices such as the Meta Quest 3.
- Open Access
- 13 June 2024
- Published by Association for Computing Machinery (ACM) in Proceedings of the ACM on Networking
- Vol. 2 (CoNEXT2) , 1-19
Twist: A Multi-site Transmission Solution for On-demand Video Streaming
Video traffic has witnessed exponential growth in recent years. As the cost optimization space of Content Delivery Network (CDN) has reached a plateau, content providers are expanding their network infrastructure to accommodate this surge. To address the cost issue, content providers have turned to cost-effective Peer-to-Peer Content Delivery Network (PCDN) solutions. However, the utilization of limited-capability PCDN nodes may lead to a decline in overall streaming performance compared to powerful CDN servers. To bridge this performance gap, we present Twist. Twist is a receiver-driven multi-site transport that leverages multiple cost-effective PCDN nodes to achieve performance equivalent to traditional CDN solutions. It incorporates a joint flow control to prevent throughput drops that can occur when using multiple content sources, and a proactive retransmission algorithm to handle frequent packet loss. Our evaluations demonstrate that Twist improves download speed on PCDN nodes by 1.71x-2.09x compared to traditional PCDN streaming solutions. Furthermore, extensive A/B testing verifies that Twist enables PCDN to achieve comparable download performance and Quality of Experience (QoE) as CDN. Over three years of commercial deployment, Twist has already served over 300 million users and has handled 35% of the video traffic, resulting in substantial cost savings for a world-leading content provider.
- 13 June 2024
- Published by Association for Computing Machinery (ACM) in Proceedings of the ACM on Networking
- Vol. 2 (CoNEXT2) , 1-19
Open AccessTwist: A Multi-site Transmission Solution for On-demand Video Streaming
Video traffic has witnessed exponential growth in recent years. As the cost optimization space of Content Delivery Network (CDN) has reached a plateau, content providers are expanding their network infrastructure to accommodate this surge. To address the cost issue, content providers have turned to cost-effective Peer-to-Peer Content Delivery Network (PCDN) solutions. However, the utilization of limited-capability PCDN nodes may lead to a decline in overall streaming performance compared to powerful CDN servers. To bridge this performance gap, we present Twist. Twist is a receiver-driven multi-site transport that leverages multiple cost-effective PCDN nodes to achieve performance equivalent to traditional CDN solutions. It incorporates a joint flow control to prevent throughput drops that can occur when using multiple content sources, and a proactive retransmission algorithm to handle frequent packet loss. Our evaluations demonstrate that Twist improves download speed on PCDN nodes by 1.71x-2.09x compared to traditional PCDN streaming solutions. Furthermore, extensive A/B testing verifies that Twist enables PCDN to achieve comparable download performance and Quality of Experience (QoE) as CDN. Over three years of commercial deployment, Twist has already served over 300 million users and has handled 35% of the video traffic, resulting in substantial cost savings for a world-leading content provider.