Donghoon Ahn

Incoming Ph.D student at the [Berkeley Artificial Intelligence Research (BAIR) Lab](https://bair.berkeley.edu/about) at UC Berkeley.

profile_img.jpg

I am an incoming Ph.D. student in Electrical Engineering and Computer Science at the University of California, Berkeley, advised by Professor Alexei Efros. I’m interested in artificial intelligence, computer vision, generative models, and diffusion models. I am particularly fascinated by the fundamental principles of how the world works and how to integrate the inductive biases derived from them into generative models. Additionally, I have a keen interest in generating images, videos, and 3D models for artistic or practical purposes.

Before starting my Ph.D. program at UC Berkeley, I was a research intern at CVLAB at KAIST. I co-authored the Perturbed-Attention Guidance (PAG), which is increasingly used in diffusion models. If you are interested, please check out my publications below.

In my free time, I enjoy drawing, listening to music, playing the piano, cycling, and exploring new places. If you have any questions or would like to connect, feel free to reach out via email (suno.vivid@gmail.com), LinkedIn, or send me a direct message on X. I welcome your messages.

If you’re interested, feel free to check out my CV.

selected publications

  1. arXiv
    noiserefine.jpeg
    A Noise is Worth Diffusion Guidance
    Donghoon Ahn, Jiwon Kang, Sanghyun Lee, and 8 more authors
    arXiv preprint arXiv:2412.03895, 2024
  2. ECCV
    pag.jpg
    Self-Rectifying Diffusion Sampling with Perturbed-Attention Guidance
    Donghoon Ahn, Hyoungwon Cho, Jaewon Min, and 6 more authors
    ECCV, 2024
  3. NeurIPS
    d-sds.jpg
    Debiasing Scores and Prompts of 2D Diffusion for View-consistent Text-to-3D Generation
    Susung Hong*, Donghoon Ahn*, and Seungryong Kim
    NeurIPS, 2023