Jailbreaking Large Language Models
This tutorial provides an overview of recent advances in jailbreaking attacks, defenses, and evaluation methods, with particular emphasis on applications in robotics and agentic systems.
Robot Proving Grounds
An introduction to robotics algorithms for mapping, localization, planning, and control, as well as a realistic testbed to try before deploying the algorithms on a real robot. The PyBullet physics simulator is used to demonstrate these algorithms on ground wheeled robots in simulation.
Embodied AI Research Takes Center Stage at San Diego Robotics Forum
UC San Diego Today || The San Diego Robotics Forum celebrated its 10th anniversary Nov. 5, 2025 with talks from startup founders who are also alumni, entrepreneur guest speakers, and engineering faculty whose work probes the limits of what is possible in robotics and artificial intelligence. This year’s event, organized by the UC San Diego […]
Congratulations to Rose Yu, recipient of the Samsung AI Researcher of the Year award!
UC San Diego professor and TILOS affiliate faculty member Rose Yu is one of three winners of the 2025 Samsung AI Researcher of the Year award! The winners were recognized at the ninth annual Samsung AI Forum, held September 15-16, 2025 in Yongin, South Korea.
Congratulations to Tajana Rosing, recipient of the 2025 SRC Sustainable Future Award!
The Semiconductor Research Corporation Sustainable Future Award recognizes researchers whose work is driving meaningful progress toward energy-efficient, sustainable technologies—whether in manufacturing or in computing and communication systems at scale.
Interdisciplinary Teams Win Three Data Storage Awards from Western Digital
UC San Diego Today || Researchers from UC San Diego, including TILOS Foundations team member Sean Gao and affiliate Rose Yu, received three Western Digital Petabyte Innovation Quest (Peak) awards for pursuing innovative data storage projects. The research teams were each awarded 1.2 petabytes of data storage—that’s more than a million gigabytes—to advance fusion energy; […]
How Robotics, AI, and Automation are Changing Our Future
In this interview with TILOS Robotics team co-lead Henrik Christensen by the Association for Advancing Automation, Christensen offers a deep look into the future of robotics and AI, and shares groundbreaking projects in healthcare robotics, autonomous mobility, and AI-driven programming for robot hands. Dr. Christensen also emphasizes the importance of trade school training, lifelong learning, […]
How faithful and trustworthy are neuron explanations in mechanistic interpretability?
Understanding what individual units in a neural network represent is a cornerstone of mechanistic interpretability. A common approach is to generate human-friendly text explanations for each neuron to describe their functionalities—but how can we trust that these explanations are faithful reflections of the model’s actual behavior?
New algorithms enable efficient machine learning with symmetric data
MIT News || A new study led TILOS Foundations team member Stefanie Jegelka and graduate student Behroozi Tahmasebi presents the first provably efficient method for machine learning with symmetric data, which could advance neural network design for applications ranging from drug discovery to climate modeling.
Unpacking the bias of large language models
MIT News || A team of MIT researchers, including TILOS Foundations team member and associate professor Stefanie Jegelka, and postdoctoral scholar Yifei Wang, has developed a theoretical framework to study how information flows through the machine learning (ML) architecture that forms the backbone of LLMs. Their work has uncovered the root cause of “position bias” […]