Tarik Can Ozden

profile_picture.jpeg

I am a Computer Science Master’s student at the University of Illinois Urbana-Champaign. I am currently a graduate research assistant at Rehg Lab working with Founder Professor James M. Rehg.

Previously, I completed my bachelor’s in Computer Engineering at Bogazici University, Turkey. I completed a research internship at UT Austin Swarm Lab under the supervision of Prof. James Rehg and Prof. Sandeep Chinchali, and Google STEP internship at Google Lens, Zurich.

My current research interests are:

  • Multimodal agentic AI,

  • Interactive AI systems,

  • Development and safety of generative models.

Education

Master of Science in Computer Science
University of Illinois Urbana-Champaign
Graduate Research Assistant at Rehg Lab
Bachelor of Science in Computer Engineering
Bogazici University
Graduated with High Honors · Exchange student at the University of Texas at Austin
High School
Izmir Science High School
Mathematical Olympiads · Valedictorian

Experience

UT Austin Swarm Lab · Research Intern
Conducted a research internship focused on preventing image editing techniques in diffusion-based models.
Developed an optimization-free, end-to-end framework for image immunization, effectively preventing edits using diffusion models.
Google Lens · STEP Intern
Redesigned and reimplemented Answer Relations at Google Lens, Zurich.

Publications

  1. EMNLP 2026
    convdeck.png
    ConvDeck: Conversational Paper-to-Slide Generation via Stage-Specific User Feedback
    Tarik Can Ozden*, Sachidanand VS*, Furkan Horoz*, Ozgur Kara, Dilek Hakkani-Tür, Junho Kim, and James M. Rehg
    Findings of the Association for Computational Linguistics: EMNLP, 2026
  2. ECCV 2026
    arcdeck.png
    Narrative-Driven Paper-to-Slide Generation via ArcDeck
    Tarik Can Ozden*, Sachidanand VS*, Furkan Horoz*, Ozgur Kara, Junho Kim, and James M. Rehg
    European Conference on Computer Vision (ECCV), 2026
  3. ICLR 2026
    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
    The Fourteenth International Conference on Learning Representations (ICLR), 2026