Computing
Efficient mobile and AI systems.
Associate Professor · University of Pittsburgh
I lead the Pittsburgh Experimental Research Group (PERG). Our work sits at the intersection of computing, sensing, and wireless communication to enable robust, human-centered intelligent systems.
Before joining Pitt, I worked as a postdoctoral research associate at Princeton and later as a senior researcher at Microsoft. I received my PhD in Computer Science from HKUST in 2015.
Efficient mobile and AI systems.
Multi-modal sensing for human behavior and environment understanding.
Wireless digital twins for cross-link and cross-band channel modeling and optimization.
Yongjie Yang, 5th year
PhD Student, Earable Computing
Tianyu Wang, 5th year
PhD Student, System for AI
Zhiwei Ren, 4th year
PhD Student, Sensor-in-the-loop AI Agent
Vaibhav Singh, 3rd year
PhD Student, Wireless Digital Twin
Junbo Li, 3rd year
PhD Student, System for AI
Hongbo Lan, 2nd year
PhD Student, Egocentric Vision System
Daiyang Yu, 2nd year
PhD Student, System for AI
Zhenlin An
Postdoc → Assistant Professor, UGA
Tao Chen
Postdoc → Senior Researcher, Samsung Research America
Maanya Shanker
Master’s Student → Software Engineer, Bloomberg
Haoyu Li
Master’s Student → PhD Student, TAMU
Yuxuan Zhang
Master’s Student → PhD Student, TAMU
Yujing Huang
Undergraduate Researcher → PhD Student, UBuffalo
UIST '26 Be My Ears: Enabling Semantic Alerting for Earphone Wearers with Pheme.
EuroSys '26 E-Cube: Event Enhanced Efficient Video Streaming for Drones.
CVPR'26 OpenSUN3D workshop Efficient Structure-Guided 3D Physical Property Reasoning.
UbiComp '25 Acoustoscillogram: Unlocking Arterial Health Insights via Acoustic Sensing with Low-cost Wired Earphones.
SenSys '25 Toward Sensor-In-the-Loop LLM Agent: Benchmarks and Implications.
SenSys '25 LeakyFeeder: In-Air Gesture Control Through Leaky Acoustic Waves.
MobiCom '24 Exploring the Feasibility of Remote Cardiac Auscultation Using Earphones.
Predict, Reuse, and Repair: Accelerating Dynamic Sparse Attention for Long-Context LLM Decoding.
Energy-Aware Scheduling for Serverless LLM Serving on Shared GPUs.
SymptomAI: Toward a Conversational AI Agent for Everyday Symptom Assessment.
CCA Reimagined: An Exploratory Study of Large Language Models for Congestion Control.
Taming Vision Priors for Data Efficient mmWave Channel Modeling.
5427 Sennott Square, Pittsburgh, PA.