Hossein Shakibania profile

Hello! I'm

Hossein Shakibania

MS Student in AIML

TU Darmstadt

  • mail
  • googleScholar
  • X
  • LinkedIn
  • github
  • cv

I'm an MS student in Artificial Intelligence and Machine Learning at Technical University of Darmstadt and a student research assistant at Multimodal AI Lab, led by Prof. Marcus Rohrbach and Prof. Anna Rohrbach. I completed my BS in Computer Engineering in 2024 at Bu-Ali Sina University.

I'm interested in multimodal generative models and their personalization, with the goal of making visual content creation more controllable, creative, and expressive.

Feel free to reach out. I'm open to collaborations!

Side quests:
🍳🎬🎶

News

Jun 2026

One co-first author paper accepted to ECCV 2026! 🎉 Looking forward to presenting Obliviate -- see you in Malmö! 🇸🇪

Apr 2026

I received the ELIZA Research-Oriented Master's Scholarship from Zuse School ELIZA! 🎉

Nov 2025

Ranked 🥇 1st in the Rayan International AI Contest! The technical report of our solutions is available on arXiv.

Apr 2025

I joined Multimodal AI Lab as a Student Research Assistant @ TU Darmstadt.

Publications

For a complete list of my research works, see my Google Scholar.

ECCV · 2026

Obliviate: Erasing Concepts from Autoregressive Image Generation Models

arXiv · 2026

Token by Token, Compromised: Backdoor Vulnerabilities in Unified Autoregressive Models

arXiv (soon) · 2026

The Poisoned Conversation: Privacy-Leaking Watermarks in Unified Multimodal Models

Digital Signal Processing · 2024

CDAN: Convolutional dense attention-guided network for low-light image enhancement

Experience

Student Research Assistant

Multimodal AI Lab

Work Study

Apr 2025 - Present

Under the supervision of Prof. Anna Rohrbach.

ECCV 2026: Proposed Obliviate, a concept erasure method for autoregressive image generation models.

Undergraduate Student Researcher

Robot Intelligence & Vision Lab

Work Study

May 2023 - June 2024

Under the supervision of Prof. Hassan Khotanlou.

Worked on low-level vision tasks, such as low-light image enhancement and cloud removal from satellite imagery.

Undergraduate Student Researcher

Intelligent Systems & Machine Learning Lab

Work Study

October 2022 - April 2023

Under the supervision of Prof. Muharram Mansoorizadeh.

Proposed a deep learning framework for detecting and grading diabetic retinopathy.

Highlights

Honors & Achievements

April 2026

ELIZA Research-Oriented Master's Scholarship, Awarded to Exceptional Students in AI and Machine Learning

November 2025

1st Place in the Rayan International AI Contest, Tehran, Iran

March 2024

Ranked 1st in the Department of Computer Engineering, Class of 2024, Bu-Ali Sina University

January 2024

Outstanding Undergraduate Student for Academic Excellence, Bu-Ali Sina University

Talks & Presentations

TU Darmstadt · Summer 2025

High-Resolution Image Synthesis with Latent Diffusion Models