Tarik Can Ozden

profile_picture.jpeg

Hi! I am a senior Computer Engineering Student at Bogazici University, Turkey.

I am currently interested in the development of advanced generative models with improved performance, safety, controllability and explainability.

Previously, I completed research internships in UIUC Rehg Lab and UT Austin Swarm Lab under the supervision of Founder Prof. James Rehg and Prof. Sandeep Chinchali. In my research, I studied safety challenges in generative models, developing techniques to mitigate risks and prevent malicious use. You can check out our paper here.

Education

BSc: Computer Engineering at Bogazici University
(Exchange Student at the University of Texas at Austin)
2020 - Present
High School: Izmir Science High School
(Mathematical Olympiads, Valedictorian)
2016 - 2020


Experience

UIUC Rehg Lab & UT Austin Swarm Lab: Research Intern
Conducting a research internship focused on preventing malicious image editing techniques in diffusion-based models.
Developed an optimization-free, end-to-end framework for image immunization, effectively preventing malicious edits in diffusion models.
May 2024 - Present


Google Lens: STEP Intern
Redesigned and reimplemented Answer Relations at Google Lens, Zurich.
July - Oct 2023



Publications

  1. In Submission
    diffvax.png
    Optimization-Free Image Immunization Against Diffusion-Based Editing
    Tarik Can Ozden*, Ozgur Kara*, Oguzhan Akcin, Kerem Zaman, Shashank Srivastava, Sandeep P. Chinchali, and James M. Rehg
    In Submission, 2024