Zikun Liu, PhD Student, UIUC

Zikun Liu 刘子坤

zikunliu [at] illinois [dot] edu

I am currently an ML research scientist. I obtained my computer science PhD degree at University of Illinois at Urbana-Champaign (UIUC) advised by Prof. Deepak Vasisht in May 2025.

Prior to UIUC, I obtained my B.E. degree in Electronic Engineering and Information Science, from School of the Gifted Young, University of Science and Technology of China (USTC) in June 2020.

My research involves designing practical and robust AI for real-world networked systems.

My research was awarded Qualcomm Innovation Fellowship'22.

GitHub / LinkedIn / Twitter / Google Scholar

News

May 2023: Started an internship at Microsoft Research for Summer 2023
Dec 2022: RAFA is accepted to NSDI 2023.
Nov 2022: BatMobility is accepted to Mobicom 2023.
Sep 2022: Awarded Qualcomm Innovation Fellowship 2022!

Selected Publications

	Counting How the Seconds Count: Understanding Algorithm-User Interplay in TikTok via ML-driven Analysis of Video Content Maleeha Masood, Shreya Kannan, Zikun Liu, Deepak Vasisht, Indranil Gupta To Appear in CHI 2026 paper We presents the Vision Language Model (VLM)-driven framework to analyze TikTok video recommendation system at scale
	Energy-aware Traffic Shaping for Cellular Radio Access Networks Zikun Liu, Seoyul Oh, Bill Tao, Yaxiong Xie, Anuj Kalia, and Deepak Vasisht NINeS 2026 code / paper We conducted the first power measurement on cellular base station with fine-grained RAN information(DCI), and designed a traffic shaping algorithm to reduce real-world power consumption without affecting user experience
	Vivisecting Starlink Throughput: Measurement and Prediction Zikun Liu, Gabriella Xue, Sarah Tanveer, and Deepak Vasisht CONEXT 2025 code / paper We conducted the first large-scale starlink throughput measurement with fine-grained satellite information, and designed a time series ML model to accurately predict the throughput.
	Exploring Practical Vulnerabilities of Machine Learning-based Wireless Systems Zikun Liu, Changming Xu, Gagandeep Singh, and Deepak Vasisht USENIX NSDI 2023 code / paper / talk We demonstrate real-world adversarial attacks on 5G machine learning systems.
	FIRE: Enabling Reciprocity for FDD MIMO Systems Zikun Liu, Gagandeep Singh, Chenren Xu, Deepak Vasisht ACM Mobicom 2021 paper / slides / talk Generative model(VAE)-based wireless channel estimation in real-world
	BatMobility: Flying Without Seeing for Lightweight Unmanned Aerial Vehicles Emerson Sie, Zikun Liu, and Deepak Vasisht ACM Mobicom 2023 video / paper / code Vision neural network + Doppler flow enable robust UAV autonomy in adverse environments lacking visual or geometric features.
	RF-Protect: Privacy against Device-Free Human Tracking Jayanth Shenoy, Zikun Liu, Bill Tao, Zach Kabelac, and Deepak Vasisht ACM SIGCOMM 2022 paper / video / code / News Conditional GAN generates fake human trajectories to protect user location privacy in wireless sensing
	One Protocol to Rule Them All: Deep Reinforcement Learning Aided MAC for Wireless Network-on-Chips Suraj Jog, Zikun Liu, Antonio Franques, Vimuth Fernando, Haitham Hassanieh, Sergi Abadal, Josep Torrellas USENIX NSDI 2021 paper / slides / talk Reinforcement learning-based MAC protocol for inter-node communication scheduling.

Other Publications

Poster: Millimeter Wave Wireless Network on Chip Using Deep Reinforcement Learning
Suraj Jog, Zikun Liu, Antonio Franques, Vimuth Fernando, Haitham Hassanieh, Sergi Abadal, Josep Torrellas
SIGCOMM'20 Student Research Competition Winner
ACM SIGCOMM 2020
paper / poster

Importance-Aware Filter Selection for Convolutional Neural Network Acceleration
Zikun Liu, Zhen Chen, Weiping Li.
2019 IEEE International Conference on Visual Communications and Image Processing (VCIP 2019), Australia.
paper

Experience

Peking University (Mar. 2021 - Jul.2021)
Research Intern, mentored by Prof. Chenren Xu

ETH Zürich (Nov. 2020 - Feb.2021)
Academic Guest, mentored by Prof. Martin Vechev

University of Illinois at Urbana-Champaign (Jul. 2019 - Nov. 2019)
Research Intern, advised by Prof. Haitham Hassanieh

University of Science and Technology of China (Nov. 2018 - May. 2019)
Research Assistant, advised by Prof. Weiping Li

University of Florida (June. 2018 - Jul. 2018)
Intern student, Summer deep learning workshop

Visitors: , Credits