Yuanhao Ban

prof_pic.jpg

banyh2000 at gmail.com

I’m a second-year Ph.D. student in UCLA Department of Computer Science advised by Professor Cho-Jui Hsieh. I have the fortune to work with Ruochen Wang, Prof. Boqing Gong, Prof. Minhao Cheng, and Prof Tianyi Zhou.

I graduated from Tsinghua University in 2023 with a degree in Eletronic Engineering. At Tsinghua, I was fortunate enough to do research at TSAIL Lab, where I worked with Dr Yinpeng Dong and Prof. Jun Zhu.

I am broadly interested in designing controllable generative models that align with human intentions and societal values, which I believe are fundamental to achieving AGI. I enjoy conducting analytical research on the mechanisms of these generative models, particularly focusing on the role of negative prompts and initial noise in image generation models, as well as the impact of positional embeddings in video diffusion models.

Here is my CV. This site was last updated on December 17th, 2024.

news

Aug 15, 2024 Invited talk on When And How Do Negative Prompts Take Effect in Dr Boqing Gong’s Group. Thanks Boqing for inviting me!
Jul 01, 2024 When And How Do Negative Prompts Take Effect got accepted by ECCV 2024!
Jun 23, 2024 I am excited to begin my internship at Snap research, where I will be working with Dr. Aliaksandr Siarohin, Dr. Yuwei Fang and Dr. Sergey Tulyakov. Our research topic focuses on motion control in video generative diffusion models.

selected publications

  1. The Crystal Ball Hypothesis in diffusion models: Anticipating object positions from initial noise
    Yuanhao Ban, Ruochen Wang, Tianyi Zhou, and 3 more authors
    arXiv preprint arXiv:2406.01970, 2024
  2. Understanding the Impact of Negative Prompts: When and How Do They Take Effect?
    Yuanhao Ban, Ruochen Wang, Tianyi Zhou, and 3 more authors
    In European Conference on Computer Vision, 2025
  3. Pre-trained adversarial perturbations
    Yuanhao Ban, and Yinpeng Dong
    Advances in Neural Information Processing Systems, 2022