Intro

I am Yuan Yao, a researcher and engineer in generative AI, focusing on diffusion models, visual generation, and multimodal learning. I earned my B.S. from the ACM Honors Class at Shanghai Jiao Tong University and am currently pursuing dual M.S. degrees in Computer Science and Data Science at University of Rochester.

My experience spans both academia and industry. Over my graduate studies, I have three years of Ph.D.-level research experience with publications at top-tier AI conferences. In industry, I build large-scale systems for image and video generation and have developed strong engineering expertise in foundation-model pipelines.

I believe research should serve real impact rather than exist for its own sake. My goal is to create generative models that not only advance academic understanding but also enable meaningful, practical, and creative applications in real world.

[CV]

Experience

Adobe Research · Research Scientist/Engineer Intern
2025/06 - Now · San Jose, CA
Video super-resolution and GRPO-based diffusion refinement
Adobe Research · Research Scientist/Engineer Intern
2024/06 - 2025/04 · San Jose, CA
Linear-complexity foundation model for text-to-image generation
OPPO US Research Center · Machine Learning Research Intern
2023/06 - 2023/12 · Bellevue, WA
Multi-stage video diffusion foundation model
University of Rochester · Ph.D.-Level Research Assistant
2022/09 - 2025/08 · Rochester, NY
Diffusion models, Generative AI, Multi-modal learning, LLMs
Shanghai AI Lab · Research Intern
2022/06 - 2023/11 · Shanghai, China
3D pre-training and multimodal representation learning

Research

2025

PixPerfect: Seamless Latent Diffusion Local Editing with Discriminative Pixel-Space Refinement
Yuan Yao*, Haitian Zheng*, Yongsheng Yu, Yuqian Zhou, Zhe Lin, Jiebo Luo
Neural Information Processing Systems (NeurIPS), 2025
[paper]

2024

Diffusion Transformer-to-Mamba Distillation for High-Resolution Image Generation
Yuan Yao, Yicong Hong, Difan Liu, Mai Long, Jiebo Luo, Feng Liu
The British Machine Vision Conference (BMVC), 2025, oral
[paper]

Pushing the Boundaries of State Space Models for Image and Video Generation
Yicong Hong, Long Mai, Yuan Yao, Feng Liu
[paper] [homepage]

Towards Open Domain Text-Driven Synthesis of Multi-Person Motions
Mengyi Shan, Lu Dong, Yutao Han, Yuan Yao, Tao Liu, Ifeoma Nwogu, Guo-Jun Qi, Mitch Hill
The European Conference on Computer Vision (ECCV), 2024
[paper] [homepage]

2023

Prompt Image to Life: Training-free Text-driven Image-to-video Generation
Jinxiu Liu, Yuan Yao, Bingwen Zhu, Fanyi Wang, Weijian Luo, Jingwen Su, Yanhao Zhang, Yuxiao Wang, Liyuan Ma, Qi Liu, Jiebo Luo, Guo-Jun Qi
[paper]

sDREAMER: Self-distilled Mixture-of-Modality-Experts Transformer for Automatic Sleep Staging
Jingyuan Chen, Yuan Yao, Mie Anderson, Natalie Hauglund, Celia Kjaerby, Verena Untiet, Maiken Nedergaard, Jiebo Luo
IEEE International Conference on Digital Health (ICDH), 2023, Best Student Paper Award
[paper]

Beyond Object Recognition: A New Benchmark towards Object Concept Learning
Yonglu Li, Yue Xu, Xinyu Xu, Xiaohan Mao, Yuan Yao, Siqi Liu, Cewu Lu
International Conference on Computer Vision (ICCV), 2023
[paper] [homepage]

2022

3D Point Cloud Pre-training with Knowledge Distillation from 2D Images
Yuan Yao, Yuanhan Zhang, Zhenfei Yin, Jiebo Luo, Wanli Ouyang, Xiaoshui Huang
ICME, 2024
[paper]

Unsupervised Learning through Shape Modeling in Medical Image Segmentation
Yuan Yao, Fengze Liu, Zongwei Zhou, Yan Wang, Wei Shen, Alan Yuille, Yongyi Lu
Medical Imaging with Deep Learning (MIDL), 2022
[paper] [code]