Faculty Earn CyLab Seed Funding

Monday, January 27, 2025

This year, CyLabOpens in new window has awarded more than $400K in seed funding to 16 CMU students, faculty, and staff members from five departments at the university. The funding was awarded on the projects’ intellectual merit, originality, potential impact, and fit towards the Security and Privacy Institute’s priorities.

One of the top priorities this year was funding projects related to security and privacy of robotics and autonomous systems, an area that CyLab is growing in collaboration with CMU’s Robotics Institute and other departments throughout the university. The Carnegie Bosch Institute (CBI)Opens in new window provided partial funding for two of the funded robotics projects.

“The seed projects we are funding this year explore approaches to mitigating misinformation, systems security, security and privacy for robotics, LLMs, and public policy issues,” said Lorrie CranorOpens in new window, director of CyLab, and professor in Carnegie Mellon’s School of Computer ScienceOpens in new window and Engineering and Public PolicyOpens in new window Department.

“CyLab partners play a key role in transforming security and privacy research into practical solutions,” said Michael LisantiOpens in new window, CyLab's senior director of partnerships. “Their support enables us to pursue ambitious projects that expand technical boundaries and address societal needs.”

The awards selection committee comprised CyLab-affiliated faculty, who prioritized several factors when making their selections, including collaborations that include junior faculty and between CyLab faculty in multiple departments, seed projects that are good candidates for follow-up funding from government or industry sources, and non-traditional projects that may be difficult to fund through other sources, among other considerations.

 Projects

A Framework for Privacy-Aware Design of the Imaging Pipeline

ROSeMont: Security Monitoring and Adaptation for ROS-based Robots

Secure and Safe FM-based Robotics using Constrained Decoding

Systematizing Privacy in Robotics

An LLM-powered Social Laboratory for Mitigating Misinformation Spread, Polarization, and Social-media Induced Violence

Anonymous Remote US ID Verification and When to Use It

Finding Date and Time Vulnerabilities with AI-Powered Differential Fuzzing

Scale-Out Encrypted LLMs on GPUs

What Can Microarchitectural Weird Machines Do?