Avatar of Ziwei Song

Ziwei Song

4th-year Ph.D. Student in Computer Science
Hong Kong SAR

About Me

I am a 4th-year Ph.D. student in the Department of Computer Science at City University of Hong Kong, advised by Prof. Nan Guan, and a member of CityU MLab. My research sits at the intersection of real-time systems and AI: making large-model inference — especially mixture-of-experts — predictable under tight memory and latency budgets, and building autonomous driving systems that can be reconfigured on the fly. More recently, I have been exploring real-time scheduling of LLM agents and LLM agents for OS-kernel formal verification. I have also worked on real-time task execution in ROS 2 robotic systems.

Prior to my Ph.D., I received my B.S. and M.S. degrees in Software Engineering from the University of Electronic Science and Technology of China in 2019 and 2022.

Real-Time Scheduling for MoE Inference Real-Time LLM Agent Scheduling ROS 2 Real-Time Robotics LLM Agents for OS Kernel Formal Verification Autonomous Driving Systems Edge Intelligence
Full CV

// the real-time atelier

Deadline Express

A brass-and-steam locomotive runs an MoE engine on a fixed memory budget. Keep the express on time.

WARMING UP
Full steam
4 / 8 in memory
tokens→router→top-2→experts —
p99 —

Hover or tap the underlined labels — including the platform automaton — to see the paper behind each part.

Instruction
Policy
p99— deadline miss— expert hit— batch—
Earlier work

Try Brake for signal with 2 experts at 30 ms — or switch to LRU and watch the tail.

What am I looking at?

Each control tick sends a batch of tokens from the locomotive to a mixture-of-experts engine. Only a few experts fit in the engine; loading an absent expert adds latency. If the total exceeds the control deadline, the express hesitates.

LRU evicts the least recently used expert. RT-MoE protects hot experts and uses free slots for prefetch.

The line outside the cab is also the research route: UESTC (the DNN years), CityU Shenzhen × Hon Hai (the Autoware / ATER autonomous truck years), and CityU MLab (the real-time MoE years). The locomotive stops briefly at each station and checkpoint.

Inspired by Autoware.Flex (RTCSA 2025) and RT-MoE (RTSS 2026). This is an illustrative simulation, not a benchmark.

  • City University of Hong Kong
    Department of Computer Science
    Ph.D. in Computer Science
    Jun. 2023 - Present
  • University of Electronic Science and Technology of China
    School of Information and Software Engineering
    M.S. in Software Engineering
    Sep. 2019 - Jun. 2022
  • University of Electronic Science and Technology of China
    School of Information and Software Engineering
    B.S. in Software Engineering
    Sep. 2015 - Jun. 2019
  • City University of Hong Kong Shenzhen Research Institute · Hon Hai Research Institute
    Autonomous Driving Light Truck Development · with Prof. Nan Guan and Prof. Jianping Wang
    Research Assistant
    Jul. 2022 - May 2023
  • Chengdu Taiji Bear Technology Co., Ltd.
    FPGA Group
    Research Intern
    Jul. 2018 - Apr. 2019
  • International Technological University
    Embedded System Group
    Research Intern
    Jul. 2016 - Aug. 2016
  • IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems (IEEE TCAD)
    Invited Reviewer
    2026
  • ACM Computing Surveys (CSUR)
    Invited Reviewer
    2026
  • Outstanding Graduate, Sichuan Province
    2022
  • Outstanding Graduate, University of Electronic Science and Technology of China
    2022
  • National Scholarship, Ministry of Education
    2021
2026
RT-MoE accepted to IEEE RTSS 2026. Read more
Jan 01
2025
Autoware.Flex and ATER presented at RTCSA 2025. Read more
Aug 20
2023
Started my Ph.D. at City University of Hong Kong, advised by Prof. Nan Guan.
Jun 01
RTSS
RT-MoE: Real-Time Expert Management for Memory-Constrained MoE Decoding

Ziwei Song, Ruoxiang Li, Fei Guan, Jen-Ming Wu, Jason Xue, Nan Guan

IEEE Real-Time Systems Symposium (RTSS) 2026 Accepted

Real-time expert management for mixture-of-experts decoding under tight memory budgets.

RT-MoE: Real-Time Expert Management for Memory-Constrained MoE Decoding

Ziwei Song, Ruoxiang Li, Fei Guan, Jen-Ming Wu, Jason Xue, Nan Guan

IEEE Real-Time Systems Symposium (RTSS) 2026 Accepted

Real-time expert management for mixture-of-experts decoding under tight memory budgets.

Autoware.Flex: Human-Instructed Dynamically Reconfigurable Autonomous Driving SystemsRTCSA
Autoware.Flex: Human-Instructed Dynamically Reconfigurable Autonomous Driving Systems

Ziwei Song, Mingsong Lv, Tianchi Ren, Chun Jason Xue, Jen-Ming Wu, Nan Guan

IEEE 31st International Conference on Embedded and Real-Time Computing Systems and Applications (RTCSA) 2025 Conference

A human-instructed, dynamically reconfigurable autonomous driving system built on the Autoware stack.

Autoware.Flex: Human-Instructed Dynamically Reconfigurable Autonomous Driving Systems

Ziwei Song, Mingsong Lv, Tianchi Ren, Chun Jason Xue, Jen-Ming Wu, Nan Guan

IEEE 31st International Conference on Embedded and Real-Time Computing Systems and Applications (RTCSA) 2025 Conference

A human-instructed, dynamically reconfigurable autonomous driving system built on the Autoware stack.

ATER: Adaptive Task Execution Rate Regulation for Enhanced Real-Time Performance in ROS 2RTCSA
ATER: Adaptive Task Execution Rate Regulation for Enhanced Real-Time Performance in ROS 2

Ruoxiang Li*, Ziwei Song*, Mingsong Lv, Jen-Ming Wu, Chun Jason Xue, Jianping Wang, Nan Guan (* equal contribution)

IEEE 31st International Conference on Embedded and Real-Time Computing Systems and Applications (RTCSA) 2025 Conference

Regulates ROS 2 task execution rates to improve real-time performance in robot software stacks.

ATER: Adaptive Task Execution Rate Regulation for Enhanced Real-Time Performance in ROS 2

Ruoxiang Li*, Ziwei Song*, Mingsong Lv, Jen-Ming Wu, Chun Jason Xue, Jianping Wang, Nan Guan (* equal contribution)

IEEE 31st International Conference on Embedded and Real-Time Computing Systems and Applications (RTCSA) 2025 Conference

Regulates ROS 2 task execution rates to improve real-time performance in robot software stacks.

Critical Path-Based Backdoor Detection for Deep Neural NetworksTNNLS
Critical Path-Based Backdoor Detection for Deep Neural Networks

Wei Jiang, Xiangyu Wen, Jinyu Zhan, Xupeng Wang, Ziwei Song, Chen Bian

IEEE Transactions on Neural Networks and Learning Systems (TNNLS) 2024 Journal

A critical-path-based detector that uses DNN interpretability to identify backdoor attacks across models.

Critical Path-Based Backdoor Detection for Deep Neural Networks

Wei Jiang, Xiangyu Wen, Jinyu Zhan, Xupeng Wang, Ziwei Song, Chen Bian

IEEE Transactions on Neural Networks and Learning Systems (TNNLS) 2024 Journal

A critical-path-based detector that uses DNN interpretability to identify backdoor attacks across models.

the paper that started it all

Semi-Federated Scheduling of Parallel Real-Time Tasks on Multiprocessors

Xu Jiang · Nan Guan · Xiang Long · Wang Yi

IEEE Real-Time Systems Symposium (RTSS) 2017 · DOI 10.1109/RTSS.2017.00015

In my first year as a master's student, I came across this paper by Nan Guan and his co-authors. I still remember how elegant the idea felt: a heavy DAG task with processing capacity x + δ is granted only x dedicated processors, while the remaining δ part is scheduled together with light tasks on shared processors. It is still one of my favourite papers in real-time scheduling. Years later I never became a pure real-time scheduling theorist, but I am still working every day toward becoming a real-time systems researcher — and this paper is still the standard I measure my own work against.

Read the paper at IEEE
Comics & Drama
Rick & Morty Family Honour One Piece Saint Seiya: The Hades Chapter
Mystery Novels
Agatha Christie John Dickson Carr Ellery Queen Arthur Conan Doyle Harlan Coben
Everything comes from the Liu & Layland boundary.