The Cake Lab

About Us

Hello! Welcome to The Cake Lab. We are a research group from the Department of Computer Science at Worcester Polytechnic Institute. Our group focuses on systems, networking, and security. For examples, some of our ongoing projects involve optimizing the performance and security of critical and emerging systems, including distributed systems, cloud and mobile applications, and embedded systems.

Our work is generously supported by the National Science Foundation and Google Cloud.


See All


Tian Guo

Assistant Professor

Robert Walls

Assistant Professor

Craig Shue

Associate Professor

Lorenzo De Carli

Assistant Professor

Mark Claypool


Craig Wills


Dan Dougherty


PhD Students

Sam Ogden

Mobile deep inference

Shijian Li

Distributed training

Yiyang Zhao

Neural architecture search

Xin Dai

Mobile-aware DNN (co-advise with Xiangnan Kong)

Guin Gilman

GPU memory management

Graduate Students

Yiqin Zhao

Mobile augmented reality

Jean-Baptiste Truong

Secure mobile deep learning

Undergraduate Students

Jake Grycel

Systems security

Yang Gao

Federated learning

Jose Li

Federated learning

Justin Aquilante

Mobile augmented reality

Tyler Jones

Mobile augmented reality


Federated learning

Scott Grubrud

Mobile deep inference


See All

DNN Model Execution Caching

The *Ripcord* project proposes new research for improving the performance of deep learning model serving. Show More

Confidential and Private Deep Learning on End-user Devices

The CAPR-DL project is exploring new techniques for providing on-device model confidentiality and user privacy. Show More

Mobile-aware Cloud Resource Management

The MOBILESCALE project proposes new research on resource management for mobile workload that differs significantly from traditional cloud workload. Show More

Embedded Systems Security

The embedded systems security project aims to protect critical embedded devices with techniques that lie at the intersection of hardware and software. Show More

Efficient Distributed Deep Learning

The *Cornucopia* project aims at identifying and mitigating the performance bottlenecks of distributed deep learning from both systems and machine learning perspective. Show More

Efficient Mobile Deep Inference

The MODI project proposes new research in designing and implementing a mobile-aware deep inference platform that combines innovations in both algorithm and system optimizations. Show More

Selected Publications

See All

Recurrent Networks for Guided Multi-Attention Classification

Xin Dai, Xiangnan Kong, Tian Guo, John Lee, Xinyue Liu, Constance Moore

ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD'20)


Silhouette: Efficient Protected Shadow Stacks for Embedded Systems

Jie Zhou, Yufei Du, Lele Ma, Zhuojia Shen, John Criswell, Robert J. Walls

USENIX Security Symposium


QuRate: Power-Efficient Mobile Immersive Video Streaming

Nan Jiang, Yao Liu, Tian Guo, Wenyao Xu, Viswanathan Swaminathan, Lisong Xu, and Sheng Wei

ACM Multimedia Systems Conference 2020 (MMSys'20)